Dual Programmed Cell Death Pathways Induced by p53 Transactivation Overcome Resistance to Oncolytic Adenovirus in Human Osteosarcoma Cells

Osteosarcoma is one of the most common malignant tumors in young children (1, 2). Current treatment strategies, which consist of multi-agent chemotherapy and aggressive surgery, have significantly improved the cure rate and prognosis of patients with osteosarcoma. In fact, over the past 30 years, the 5-year survival rate has increased from 10% to 70% (3–5). Even in patients with osteosarcoma with metastases at diagnosis, the 5-year survival rate has reached 20% to 30% in response to chemotherapy and surgical removal of primary and metastatic tumors (6). However, treatment outcomes for patients with osteosarcomas have further improved over the last few years. Therefore, the development of novel therapeutic strategies is required to improve the clinical outcomes in patients with osteosarcomas.

Tumor-specific replication-competent oncolytic viruses are being developed as novel anticancer therapy, in which the promoters of cancer-related genes are used to regulate virus replication in a tumor-dependent manner. More than 85% of all human cancers express high telomerase activity to maintain the length of the telomeres during cell division, whereas normal somatic cells seldom show this enhanced telomerase activity (7, 8). Telomerase activity has also been detected in 44% to 81% of bone and soft-tissue sarcomas (9, 10). Telomerase activation is closely correlated with the expression of the human telomerase reverse transcriptase (hTERT) gene (11). On the basis of these data, we previously developed a telomerase-specific replication-competent oncolytic adenovirus OBP-301 (Telomelysin) in which the hTERT gene promoter drives the expression of the E1A and E1B genes (12). A phase I clinical trial of OBP-301, which was conducted in the United States on patients with advanced solid tumors, indicated that OBP-301 was well tolerated by patients (13). Recently, we reported that OBP-301 efficiently killed human bone and soft-tissue sarcoma cells (14, 15). However, some osteosarcoma cell lines were not sensitive to the antitumor effect of OBP-301. Therefore, to efficiently eliminate tumor cells with OBP-301, its antitumor effects need to be enhanced.

Cancer gene therapy is defined as the treatment of malignant tumors via the introduction of a therapeutic tumor suppressor gene or the abrogation of an oncogene. The tumor suppressor p53 gene has an attractive tumor suppressor profile as a potent therapeutic transgene for induction of cell-cycle arrest, senescence, apoptosis, and autophagy (16). Dual cell death pathways, such as apoptosis and autophagy, induced by p53 transactivation are mainly involved in the suppression of tumor initiation and progression. However, among the p53 downstream target genes, p21, which is most rapidly and strongly induced during the DNA damage response, mainly induces cell-cycle arrest through suppression of apoptotic and autophagic cell death pathways (17, 18). Thus, p21 suppression may be a more effective strategy for the induction of apoptotic and autophagic cell death pathways in tumor cells, particularly when the tumor suppressor p53 gene is overexpressed in tumor cells in response to cancer gene therapy.

A p53-expressing replication-deficient adenovirus (Ad-p53, Advexin) has previously been reported to induce an antitumor effect in the in vitro and in vivo settings (19, 20) as well as in some clinical studies (21–24). We recently reported that combination therapy with OBP-301 and Ad-p53 resulted in a more profound antitumor effect than monotherapy with either OBP-301 or Ad-p53 (25). Moreover, we generated armed OBP-301 expressing the wild-type p53 tumor suppressor gene (OBP-702) and showed that OBP-702 suppressed the viability of various types of epithelial malignant cells more efficiently than did OBP-301 (26). OBP-702 induced a more profound apoptotic cell death effect than Ad-p53, likely via adenoviral E1A-mediated suppression of anti-apoptotic p21 in human epithelial malignant cells. However, it remained unclear whether OBP-702 efficiently induces an antitumor effect in human nonepithelial malignant cells, including osteosarcomas.

In the present study, we investigated the in vitro cytopathic efficacy of the p53-expressing telomerase-specific replication-competent oncolytic adenovirus, OBP-702, in human osteosarcoma cells, and we compared the induction level of apoptotic and autophagic cell deaths in OBP-301–resistant human osteosarcoma cells infected with OBP-301, OBP-702, and Ad-p53. The molecular mechanism by which OBP-702 mediates induction of apoptosis and autophagy was also investigated. Finally, the in vivo antitumor effect of OBP-702 was evaluated using an orthotopic OBP-301–resistant human osteosarcoma xenograft tumor model.

The human osteosarcoma cell lines, HOS and SaOS-2, were kindly provided by Dr. Satoru Kyo (Kanazawa University, Ishikawa, Japan). These cells were propagated as monolayer cultures in Dulbecco's Modified Eagle's Medium. The human osteosarcoma cell line, U2OS, was obtained from the American Type Culture Collection and was grown in McCoy's 5a Medium. The human osteosarcoma cell line, MNNG/HOS, was purchased from DS Pharma Biomedical and was maintained in Eagle's Minimum Essential Medium containing 1% nonessential amino acids. All media were supplemented with 10% FBS, 100 U/mL penicillin, and 100 mg/mL streptomycin. The normal human lung fibroblast (NHLF) cell line, NHLF, was obtained from TaKaRa Biomedicals. NHLF cells were propagated as monolayer culture in the medium recommended by the manufacturer. Although cell lines were not authenticated by the authors, cells were immediately expanded after receipt and stored in liquid N₂. Cells were not cultured for more than 5 months following resuscitation. The cells were maintained at 37°C in a humidified atmosphere with 5% CO₂.

The recombinant telomerase-specific replication-competent adenovirus OBP-301 (Telomelysin), in which the promoter element of the hTERT gene drives the expression of E1A and E1B genes, was previously constructed and characterized (12, 27). For OBP-301–mediated induction of exogenous p53 gene expression, we recently generated OBP-702, in which a human wild-type p53 gene expression cassette was inserted into the E3 region (Supplementary Fig. S1; ref. 26). Ad-p53 is a replication-defective adenovirus serotype 5 vector with a p53 gene expression cassette at the E1 region (19, 20). Recombinant viruses were purified by ultracentrifugation using cesium chloride step gradients, their titers were determined by a plaque-forming assay using 293 cells and they were stored at −80°C.

Cells were seeded on 96-well plates at a density of 1 × 10³ cells/well 24 hours before viral infection. All cell lines were infected with OBP-702 at multiplicity of infections (MOI) of 0, 0.1, 1, 10, 50, or 100 plaque-forming units (PFU)/cell. Cell viability was determined on days 1, 2, 3, and 5 after virus infection using Cell Proliferation Kit II (Roche Molecular Biochemicals). The 50% inhibiting dose (ID₅₀) value of OBP-702 for each cell line was calculated using cell viability data obtained on day 5 after virus infection.

GFP-expressing MNNG/HOS (MNNG/HOS-GFP) cells were established by stable transfection with GFP expression plasmid using Lipofectamine LTX (Invitrogen). MNNG/HOS-GFP and NHLF cells were seeded in 35-mm glass-based dishes at a density of 1 × 10⁵ cells/dish 24 hours before infection and were infected with OBP-702 at an MOI of 10 PFU/cell for 72 hours. Phase-contrast and fluorescence time-lapse recordings were obtained to concomitantly analyze cell morphology and GFP expression using an inverted FV10i confocal laser scanning microscopy (OLYMPUS).

SaOS-2 and MNNG/HOS cells, seeded in a 100-mm dish at a density of 1 × 10⁵ cells/dish, were infected with OBP-301, OBP-702, or Ad-p53 at the indicated MOIs. In contrast, SaOS-2 cells were transfected with 10 nmol/L miR-93 (Ambion), miR-106b (Ambion), or control miRNA (Ambion) 24 hours before Ad-p53 infection and infected with Ad-p53 at an MOI of 100 for 48 hours. Whole-cell lysates were prepared in a lysis buffer [50 mmol/L Tris-HCl (pH 7.4), 150 mmol/L NaCl, 1% Triton X-100] containing a protease inhibitor cocktail (Complete Mini; Roche) at the indicated time points. Proteins were electrophoresed on 6% to 15% SDS-PAGE and were transferred to polyvinylidene difluoride membranes (Hybond-P; GE Health Care). Blots were blocked with 5% non-fat dry milk in TBS-T (Tris-buffered saline and 0.1% Tween-20, pH 7.4). The primary antibodies used were: rabbit anti-PARP polyclonal antibody (pAb; Cell Signaling Technology), mouse anti-p53 monoclonal antibody (mAb; Calbiochem), mouse anti-p21^WAF1 mAb (Calbiochem), rabbit anti-E2F1 pAb (Santa Cruz Biotechnology), mouse anti-Ad5 E1A mAb (BD PharMingen), rabbit anti-microtubule-associated protein 1 light chain 3 (LC3) pAb [Medical & Biological Laboratories (MBL)], mouse anti-p62 mAb (MBL), rabbit anti-damage-regulated autophagy modulator (DRAM) pAb (Abgent), and mouse anti-β-actin mAb (Sigma-Aldrich).

To analyze the active caspase-3 expression, cells were incubated for 20 minutes on ice in Cytofix/Cytoperm solution (BD Biosciences), labeled with phycoerythrin (PE)-conjugated rabbit anti-active caspase-3 mAb (BD Biosciences) for 30 minutes, and then analyzed using FACS array (BD Biosciences).

To evaluate the sub-G₁ population, which is a apoptosis indicator, in SaOS-2 cells after virus infection, SaOS-2 cells were seeded in a 100-mm dish at a density of 1 × 10⁶ cells/dish 24 hours before viral infection and were infected with mock, OBP-301, Ad-p53, or OBP-702 at an MOI of 10 PFUs/cell for 48 hours. Cells were trypsinized and resuspended in original supernatant to ensure that both attached and nonattached cells were analyzed. Cells stained with propidium iodide were analyzed using FACS array (BD Biosciences).

To evaluate the expressions of miR-93 and miR-106b in tumor cells after OBP-702 infection, SaOS-2 and MNNG/HOS cells were seeded on 6-well plates at a density of 2 × 10⁵ cells/well 24 hours before viral infection and were infected with OBP-702 at MOIs of 0, 1, 5, 10, 50, or 100 PFU/cell. Three days after virus infection, total RNA was extracted from the cells using a miRNeasy Mini Kit (Qiagen). The concentration and quality of RNA were assessed using a Nanodrop spectrophotometer. cDNA was synthesized from 10 ng of total RNA using the TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems), and quantitative real-time reverse transcription (RT)-PCR was carried out using the Applied Biosystems StepOnePlus real-time PCR System. The expressions of miR-93 and miR-106b were defined from the threshold cycle (C_t), and relative expression levels were calculated using the |$2^{- \Delta \Delta C_{\rm t}}$| method after normalization with reference to the expression of U6 small nuclear RNA.

Animal experimental protocols were approved by the Ethics Review Committee for Animal Experimentation of Okayama University School of Medicine (Okayama, Japan). MNNG/HOS cells (5 × 10⁶ cells per site) were inoculated into the tibias of female athymic nude mice aged 6 to 7 weeks (CLEA Japan). Palpable tumors developed within 21 days and were permitted to grow to approximately 5 to 6 mm in diameter. At that stage, a 50-μL volume of solution containing OBP-702, OBP-301, or Ad-p53 at a dose of 1 × 10⁸ PFU or PBS was injected into the tumors for 3 cycles every 2 days. Tumor volume was monitored by computed tomographic (CT) imaging once a week after virus infection.

The tumor volume and formation of osteolytic lesions were evaluated using three-dimensional CT (3D-CT) imaging (ALOKA Latheta LCT-200; Hitachi Aloka Medical). The tumor volume was calculated by INTAGE Realia software (Cybernet Systems).

Tumors were fixed in 10% neutralized formalin and embedded in paraffin blocks. Sections were stained with hematoxylin/eosin and analyzed by light microscopy.

Data are expressed as means ± SD. Student t test was used to compare differences between groups. Statistical significance was defined as P < 0.05.

To evaluate the in vitro cytopathic activity of OBP-702, we used the 2 OBP-301–sensitive human osteosarcoma cells (HOS and U2OS) and the 2 OBP-301–resistant human osteosarcoma cells (SaOS-2 and MNNG/HOS) that were recently described (14). The cell viability of each cell was assessed over 5 days after infection using the XTT assay. OBP-702 infection suppressed the viability of OBP-301–sensitive and -resistant cells in dose- and time-dependent manners (Fig. 1A and B). When the ID₅₀ values of OBP-702 in all 4 human osteosarcoma cells were compared with those of OBP-301 calculated in a previous report (14), all cell lines were more sensitive to OBP-702 than to OBP-301 (Table 1). The ID₅₀ values of OBP-702 were also lower than those of Ad-p53 (Supplementary Fig. S2). However, OBP-702 did not exhibit any cytopathic effect in NHLF cells (Fig. 1B). When GFP-expressing MNNG/HOS-GFP cells were cocultured with human normal NHLF cells, OBP-702 infection showed a cytopathic effect (confirmed by observation of round-shaped morphologic changes) in MNNG/HOS-GFP cells but not in NHLF cells (Fig. 1C). These results indicate that OBP-702 was more cytopathic than OBP-301 for human osteosarcoma cells but was not cytopathic for normal human cells.

We next investigated whether OBP-702 induces more profound apoptosis when compared with OBP-301 or Ad-p53. OBP-301–resistant SaOS-2 and MNNG/HOS cells were infected with OBP-702, OBP-301, or Ad-p53, and apoptosis was assessed by Western blot and flow cytometric analyses. Western blot analysis showed that SaOS-2 cells exhibited the cleavage of PARP after infection with OBP-702 (>5 MOIs) or Ad-p53 (>50 MOIs), whereas MNNG/HOS cells had the cleavage of PARP after infection with OBP-702 (>5 MOIs) but not Ad-p53 (Fig. 2A). In contrast, OBP-301 did not induce apoptosis (data not shown). Furthermore, flow cytometric analysis showed that OBP-702 infection (10 MOIs) significantly increased the percentage of apoptotic cells with active caspase-3 when compared with Ad-p53 or OBP-301 at same doses in SaOS-2 and MNNG/HOS cells (Fig. 2B and C). Cell-cycle analysis also showed that OBP-702 (10 MOIs) induced the highest percentages of sub-G₁ population in SaOS-2 cells when compared with Ad-p53 or OBP-301 at same doses (Fig. 2D). These results suggest that OBP-702 induces increased apoptosis when compared with Ad-p53 or OBP-301 in human osteosarcoma cells.

To investigate the molecular mechanism of OBP-702–induced apoptosis in human osteosarcoma cells, we evaluated p53 expression after OBP-702 infection in SaOS-2 (p53-null) and MNNG/HOS (p53-mutant) cells in which endogenous p53 expression level was confirmed by Western blot analysis (Supplementary Fig. S3). OBP-702 efficiently induced p53 expression in SaOS-2 and MNNG/HOS cells (Fig. 3A). The level of p53 expression was higher in OBP-702–treated cells than in Ad-p53–treated cells (Fig. 3A). Despite of OBP-702–induced high p53 expression, p53 downstream target p21 protein was induced only in Ad-p53–treated cells.

To investigate the effect of exogenous p53 overexpression in virus replication, we next compared the replication abilities of OBP-702 and OBP-301 in p53-null SaOS-2 cells by measuring the relative amounts of E1A copy numbers. The E1A copy number of OBP-702 was similar to that of OBP-301 in SaOS-2 cells (Supplementary Fig. S4). These results indicate that OBP-702 efficiently induces exogenous p53 expression without affecting p21 expression and virus replication in human osteosarcoma cells.

Adenoviral E1A protein has been shown to activate E2F1 expression (28), which is a multifunctional transcription factor that regulates diverse cell fates through induction of many target genes, including small noncoding miRNAs (29). Recently, E2F1-inducible miR-93 and miR-106b have been shown to suppress p21 expression in human cancer cells (30). Therefore, we sought to investigate whether OBP-702 induces expressions of E2F1 and E2F1-regulated miRNAs (miR-93 and miR-106b). OBP-702 infection activated E2F1 expression along with E1A accumulation in SaOS-2 and MNNG/HOS cells (Fig. 3B). The expression levels of miR-93 and miR-106b were increased in association with E2F1 activation in OBP-702–infected SaOS-2 and MNNG/HOS cells (Fig. 3C). In contrast, E1A-deleted Ad-p53 infection did not increase expressions of E2F1 and E2F1-regulated miR-93 and miR-106b (data not shown). Next, we assessed whether upregulation of miR-93 and miR-106b efficiently suppresses p21 expression induced by Ad-p53–mediated p53 overexpression. Ad-p53 infection at MOIs of 10 and 100 efficiently induced p21 expression at 48 hours after infection in SaOS-2 cells (Supplementary Fig. S5). When SaOS-2 cells were infected with Ad-p53 at an MOI of 100 for 48 hours, pretransfection with miR-93, miR-106b, or both efficiently suppressed Ad-p53–induced p21 expression (Fig. 3D). Interestingly, both miR-93- and miR-106b–transfected SaOS-2 cells showed the 1.5-fold increased expression of cleaved PARP (C-PARP) in consistency with remarkable p21 downregulation when compared with those transfected with control miR. However, the expression level of C-PARP was not increased in the miR-93- or miR-106b–transfected SaOS-2 cells, although transfection with miR-93 or miR-106b moderately decreased p21 expression. These results suggest that OBP-702 suppresses p21 expression through E1A-dependent upregulation of both E2F1-inducible miR-93 and miR-106b and contributes to induction of apoptosis.

Recently, we showed that oncolytic adenovirus OBP-301 mainly induces programmed cell death in association with autophagy rather than apoptosis in human tumor cells (31). Therefore, we next investigated whether OBP-702 induces more profound autophagy than does OBP-301. Western blot analysis revealed that OBP-702 infection showed increased autophagy, which was confirmed by conversion of LC3-I to LC3-II (increased ratio of LC3-II/LC3-I) and p62 downregulation, when compared with OBP-301 in MNNG/HOS cells (Fig. 4A). Moreover, the expression level of the p53-induced modulator of autophagy, DRAM (32), was decreased after OBP-301 infection, but its expression was maintained after OBP-702 infection (Fig. 4A). As p53-mediated p21 overexpression is known to inhibit both apoptosis and autophagy (17, 18), we further evaluated whether miR-mediated p21 suppression is involved in the enhancement of p53-mediated autophagy induction. Ad-p53–induced autophagy was enhanced by miR-93- and miR-106b–mediated p21 suppression (Fig. 4B). These results suggest that OBP-702 induces more profound autophagy than does OBP-301 and that this effect occurs via p53-mediated DRAM activation and miR-mediated p21 suppression.

Finally, to assess the in vivo antitumor effect of OBP-702, we used an orthotopic MNNG/HOS tumor xenograft model. OBP-702, OBP-301, Ad-p53, or PBS were intratumorally injected for 3 cycles every 2 days. OBP-702 administration significantly suppressed tumor growth when compared with OBP-301, Ad-p53, or PBS in an orthotopic MNNG/HOS tumor model (Fig. 5A and B). 3D-CT examination revealed that OBP-702–treated tumors had less bone destruction than did OBP-301- or Ad-p53–treated tumors (Fig. 5C). On histopathologic analysis, there were large necrotic areas in OBP-702–treated tumors but not in OBP-301- or Ad-p53–treated tumors (Fig. 5D). Moreover, the expression of the cell proliferation marker, Ki67, was also decreased, especially in OBP-702–treated tumor cells (Supplementary Fig. S6). These results suggest that OBP-702 eliminates tumor tissues more efficiently when compared with OBP-301 or Ad-p53.

We previously reported that telomerase-specific replication-competent oncolytic adenovirus OBP-301 has strong antitumor activity in a variety of human epithelial and nonepithelial malignant cells (12, 14, 27). However, some human osteosarcoma cells were resistant to the cytopathic activity of OBP-301 (14). In this study, we showed that a novel p53-expressing oncolytic adenovirus, OBP-702, had increased in vitro and in vivo antitumor effects than did OBP-301 in human osteosarcoma cells (Fig. 1 and 5). OBP-702 induced increased apoptosis in association with p53 upregulation and p21 downregulation when compared with replication-deficient Ad-p53 (Fig. 2 and 3A). E1A-dependent upregulation of miR-93 and miR-106b was involved in OBP-702–mediated suppression of p21 expression (Fig. 3). Moreover, p53-mediated DRAM activation with p21 suppression enhanced oncolytic adenovirus-mediated autophagy induction (Fig. 4). Recent studies suggest that transgene-expressing armed oncolytic adenoviruses are a promising antitumor strategy for induction of oncolytic and transgene-induced cell death (33). Although p53 overexpression has been shown to enhance antitumor activity of oncolytic adenoviruses (34), the molecular mechanisms by which p53 mediates enhancement of the antitumor effect remain unclear. Recently, we reported that OBP-702 induces profound apoptosis through p53-dependent BAX upregulation and E1A-dependent p21 and MDM2 downregulation in epithelial malignant cells (26). Thus, oncolytic adenovirus-mediated p53 overexpression likely induces dual apoptotic and autophagic cell death pathways through p53-dependent BAX/DRAM activation and E1A-dependent p21/MDM2 suppression with E2F1-inducible miR-93/106b upregulation (Fig. 6).

OBP-702 efficiently suppressed the cell viability of both OBP-301–sensitive and -resistant osteosarcoma cells (Fig. 1). We previously reported that OBP-301–resistant SaOS-2 cells have no hTERT mRNA expression (Table 1), suggesting that SaOS-2 cells maintain telomere length through alternative lengthening of telomeres (ALT). As hTERT gene promoter is used for tumor-specific replication of OBP-301, ALT-type human osteosarcoma cells such as SaOS-2 cells may be resistant to OBP-301. However, ALT-type SaOS-2 cells showed similar sensitivity to OBP-702 as well as non–ALT-type MNNG/HOS cells (Fig. 1 and Table 1). These results suggest that p53 overexpression overcomes resistance to OBP-301 in ALT-type SaOS-2 cells. As the replication rate of OBP-702 was almost similar that of OBP-301 in ALT-type SaOS-2 cells (Supplementary Fig. S2), p53-induced cell death pathway would suppress the cell viability of ALT-type human osteosarcoma cells.

OBP-702–mediated p53 overexpression induced 2 types of programmed cell deaths (i.e., apoptosis and autophagy), thereby contributing to the enhancement of the antitumor effect of OBP-301 in human osteosarcoma cells (Fig. 2 and 4). As p53 downstream target p21 functions as a suppressor of apoptosis and autophagy (17, 18), p21 suppression may be a critical factor to induce dual programmed cell death pathways in response p53 overexpression. Suppression of p21 expression by genetic deletion or artificial p21 target microRNA has been shown to enhance the Ad-p53–induced apoptosis (18, 35). Inactivation of p21 by adenoviral E1A has been shown to enhance apoptosis in chemotherapeutic drug-treated human colon cancer cells that overexpress p53 (36). Genetic deletion of p21 has been also shown to induce autophagy in mouse embryonic fibroblasts treated with C(2)-ceramide or γ-irradiation (17). In contrast, p21 overexpression inhibited the Ad-p53–mediated apoptosis induction (18). Thus, E1A-mediated p21 downregulation would enhance p53-induced apoptosis and autophagy in OBP-702–infected cells.

E1A-dependent E2F1 activation and subsequent upregulation of E2F1-inducible miRNAs efficiently suppressed p21 expression, leading to the enhancement of p53-induced apoptosis and autophagy, in OBP-702–infected osteosarcoma cells (Figs. 2–4). Recent studies suggest that the cross-talk between p53 and E2F1 play a role in the regulation of diverse cell fates (37). For example, co-expression of p53 and E2F1 contributes to induction of apoptosis (38, 39). We previously showed that E2F1 enhanced Ad-p53–mediated apoptosis through p14ARF-dependent MDM2 downregulation (39) and that OBP-702 infection showed E1A-dependent MDM2 downregulation in association with apoptosis (26). Recently, E2F1 has been shown to suppress MDM2 expression by suppressing the promoter activity (40) or by inducing upregulation of miR-25/32, which targets MDM2 (41). Furthermore, E2F1-inducible miR-93/106b enhanced Ad-p53–induced apoptosis and autophagy via p21 suppression (Figs. 3D and 4B). Therefore, the cooperation between the MDM2/p53/p21 pathway and the E2F1/miRNA pathway may be involved in the induction of apoptotic and autophagic cell death in response to OBP-702.

OBP-702–mediated p53 overexpression enhanced autophagy that was induced by oncolytic adenovirus in human osteosarcoma cells. OBP-702 infection induced increased expression of DRAM and decreased expression of p62 when compared with OBP-301 (Fig. 4), suggesting that OBP-702–mediated p53 overexpression enhances autophagy through DRAM activation. We recently reported that OBP-301 induces autophagy through E1A-dependent activation of E2F1/miR-7 pathway and subsequent suppression of EGF receptor (EGFR; ref. 31). Restoration of p53 expression enhances the sensitivity to EGFR inhibitors in human cancer cells (42). Moreover, EGFR downregulation by transfection of specific antisense oligonucleotide promotes the differentiation status of human osteosarcoma U2OS cells (43). Thus, OBP-702 may induce differentiation as well as cell death through autophagy activation by DRAM upregulation and EGFR downregulation in human osteosarcoma cells.

The 3D-CT imaging system was a useful method to assess both tumor volume and bone destruction status in MNNG/HOS tumors. OBP-702–treated tumors were smaller and had less bone destruction than PBS-, Ad-p53-, or OBP-301–treated tumors (Fig. 5A and C). Recent reports have suggested that zoledronic acid suppresses tumor growth as well as osteolytic components in human osteosarcoma xenograft tumor models (44, 45). These results suggest that combination therapy with OBP-702 and zoledronic acid may be more effective and more protective against bone destruction in human osteosarcomas. Further study using a 3D-CT imaging system may provide important information about bone destruction status in osteosarcomas treated with OBP-702 and zoledronic acid.

Adenovirus-mediated p53 gene therapy exerts an antitumor effect in human osteosarcoma cells (46). However, the antitumor activity of replication-deficient Ad-p53 is limited in some human osteosarcoma cells (47). Ad-p53–mediated p53 overexpression increases the sensitivity of human osteosarcoma cells to the chemotherapeutic drugs, cisplatin and doxorubicin (48). A synergistic antitumor effect between doxorubicin and roscovitine was also associated with autophagy induction in human osteosarcoma U2OS cells (49). As OBP-702 induced more profound apoptosis and autophagy than did OBP-301 or Ad-p53 (Fig. 2 and 4), combination therapy with OBP-702 and chemotherapeutic agents may be more effective than monotherapy with OBP-702. Moreover, a recent report has shown that p53-armed replication-competent oncolytic adenovirus is a safe antitumor agent in rodents and non-human primates (50). However, for clinical application of OBP-702, it must be necessary to establish the systemic delivery method and confirm the host biologic contributions in patients with cancer. Although there are some unsolved issues, the combination of p53-armed oncolytic adenovirus and chemotherapy may provide us a promising antitumor strategy against human osteosarcoma cells.

In conclusion, we clearly showed that the p53-expressing oncolytic adenovirus OBP-702 has a much stronger antitumor effect than does OBP-301. Oncolytic adenovirus-mediated p53 gene transduction may induce dual apoptotic and autophagic cell death pathways through p53-dependent activation of cell death inducers and E1A-dependent suppression of cell death inhibitors, resulting in the enhancement of antitumor effect.

Y. Urata is President & CEO of Oncolys BioPharma, Inc., the manufacturer of OBP-301 (Telomelysin). H. Tazawa and T. Fujiwara are consultants of Oncolys BioPharma, Inc. No potential conflicts of interest were disclosed by the other authors.

Conception and design: J. Hasei, F. Uno, S. Kagawa, T. Ozaki, T. Fujiwara

Development of methodology: J. Hasei, F. Uno, S. Kagawa

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J. Hasei, H. Tazawa, S. Osaki, Y. Yamakawa, A. Yoshida, T. Onishi, T. Ozaki

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J. Hasei, H. Tazawa, T. Fujiwara

Writing, review, and/or revision of the manuscript: J. Hasei, H. Tazawa, T. Kunisada, Y. Urata, T. Fujiwara

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): H. Tazawa, T. Kunisada, Y. Hashimoto, S. Kagawa, Y. Urata, T. Ozaki

Study supervision: T. Sasaki, T. Kunisada, F. Uno, Y. Urata, T. Fujiwara

The authors thank Dr. Satoru Kyo (Kanazawa University) for providing the HOS and SaOS-2 cells and Tomoko Sueishi for her excellent technical support.

This study was supported by grants-in-aid from the Ministry of Education, Science, and Culture, Japan (T. Fujiwara) and grants from the Ministry of Health and Welfare, Japan (T. Fujiwara), and in part by the National Cancer Center Research and Development Fund (23-A-10) (T. Ozaki).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.