Adenovirus-expressed human hyperplasia suppressor gene induces apoptosis in cancer cells Free

Hyperplasia suppressor gene (HSG) is a novel gene characterized by Chen et al. as a cell proliferation suppressor (1). It was originally named hypertension-related gene-1 because it was found in cultured vascular smooth muscle cells of Wistar-Kyoto and spontaneous hypertension rats using mRNA differential display technology (2). The expression of hypertension-related gene-1 mRNA was lower in spontaneous hypertension rat than Wistar-Kyoto rat. Moreover, vascular smooth muscle cell proliferation was inhibited by hypertension-related gene-1 (3). The hypertension-related gene-1 name was changed to HSG after overexpression was found to suppress serum-evoked vascular smooth muscle cell proliferation in culture. Because rHSG shares 95.2% and 98.4% similarity with its human and mouse homologues, respectively, we cloned human HSG (hHSG) to further explore its function. hHSG is located in chromosome 1p36.3. hHSG is also named human mitofusin 2 (Mfn2), because its homologue, Mfn1, plays a vital role in mitochondrial fusion by regulating mitochondrial morphology and function in mammalian cells, yeast, and flies (4–7). When transiently expressed in mammalian cells, both variants, Mfn1 and Mfn2, induce formation of highly elongated and interconnected mitochondrial tubules. Recent studies have indicated that Bax or Bak is required for normal fusion of mitochondria into elongated tubules. Bax induces mitochondrial fusion by activating assembly of the large GTPase Mfn2 through interaction with coiled-coil domain of Mfn2 (8). As a proapoptosis protein, Bax colocalizes with Mfn2 during apoptosis (9). These evidences imply that hHSG may be also involved in apoptosis.

In this study, to investigate hHSG bioactivity, we constructed an adenovirus expression vector of hHSG (Ad5-hHSG) and found that Ad5-hHSG had a wide antitumor effect in vitro. Interestingly, it was not efficient in some diploid cell lines. In vitro and in vivo experimental data confirmed that hHSG overexpression induces cancer cell apoptosis, along with mitochondrial membrane potential (ΔΨm) reduction and release of cytochrome c. In addition, Ad5-hHSG also induced chemosensitization and radiosensitization of A549 and HT-29 cells. Our results suggest that adenovirus-mediated hHSG gene transfer may present a new therapeutic approach for cancer treatment.

A549, HeLa, HT-29, MCF-7, HEK293, normal human foreskin fibroblast cell line BJ, HL-60, K562, Raji, and U937 were obtained from American Type Culture Collection. Normal liver cell line L02 was obtained from the cell bank of Type Culture Collection of Chinese Academy of Science.

A549, HeLa, HT-29, MCF-7, L02, and HEK293 were cultured in DMEM (Life Technologies) supplemented with 10% fetal bovine serum (Hyclone), 2 mmol/L glutamine, 100 units/mL penicillin, and 100 μg/mL streptomycin. BJ was cultured in MEM medium (Life Technologies) supplemented as above. HL-60, K562, Raji, and U937 were cultured in RPMI 1640 (Life Technologies) supplemented as above. All the cells were maintained in an incubator with a humidified atmosphere of 95% air and 5% CO₂ at 37°C.

The complete coding region hHSG PCR product was subcloned into the pDC316-cytomegalovirus shuttle vector. The pDC316-hHSG DNA was recombined with backbone pAdEasy-1 in BJ5183 bacteria. Adenovirus generation, amplification, and titer were done according to the simplified system described by He et al. (10). Viral particles were purified by cesium chloride density gradient centrifugation. The control virus or the virus containing hHSG was named Ad5-Null or Ad5-hHSG, respectively.

Cell viability after infection with different adenoviral vectors was determined using the Vi-CELL TMXR cell viability analyzer (Beckman Coulter). Trypan blue–positive cells were considered dead, and the numbers of viable cells were calculated by the analyzer according to the manufacturer's instructions.

Detection of hHSG mRNA expression was carried out by reverse transcription–PCR. Cells were collected at indicated time points, and RNA was prepared using TRIzol reagent (Invitrogen Life Technologies) according to the manufacturer's protocol. Reverse transcription was done with the ThermoSCRIPT reverse transcription–PCR system (Invitrogen Life Technologies), according to the manufacturer's protocol. A total RNA of 1 μg was used for the first strand cDNA synthesis using oligo (dT) 15 primer. The complete coding region of hHSG was amplified from human tumor cells cDNA library by PCR using the forward primer 5′-ATGTCCCTGCTCTTCTCTCGATGC-3′ and reverse primer 5′-TCTGCTGGGCTGCAGGTACTGGT-3′. The PCR reaction for hHSG amplification was carried out with one cycle of 94°C for 5 min and 30 cycles of 94°C for 1 min, 60°C for 1 min, 72°C for 2 min, followed by a one cycle extension at 72°C for 7 min. For glyceraldehyde-3-phosphate dehydrogenase (internal control) primers, forward primer 5′-CCACCCATGGCAAATTCCATGGCA-3′, reverse primer 5′-TCTAGACGGCAGGTCAGGTCAGGTCCACC- 3′ were used, and PCR reaction was done at an annealing temperature of 58°C for 20 cycles. PCR products were separated via 1.0% agarose gel electrophoresis and visualized by ethidium bromide staining.

The level of hHSG protein expression, the release of cytochrome c, and cleavage of PARP were examined by immunoblotting technique. Cells were extracted using a lysis buffer (10 mmol/L HEPES, 150 mmol/L NaCl, 1 mmol/L EDTA, 1 mmol/L EGTA, 1% Triton-X100, 0.5% NP40, 0.1% SDS, and with freshly added proteinase inhibitor cocktail) for 30 min on ice. The protein concentration was quantified using the bicinchoninic acid protein assay reagent (Pierce). Equal amounts of protein (30 μg) were separated by SDS-PAGE and transferred onto nitrocellulose membranes (Amersham Pharmacia). Membranes were incubated overnight at 4°C with the appropriate primary antibodies (anti-HSG rabbit polyclonal antibody, 1:1,000, bestowed by Shigehisa Hirose, Department of Biological Sciences, Tokyo Institute of Technology; anti–cytochrome c mouse monoclonal antibody, 1:1,000, BD PharMingen; anti-PARP rabbit polyclonal antibody, 1:1,000, Cell Signaling Technology; anti-Bax rabbit polyclonal antibody, 1:1,000, Abcam). Incubated for 1 h in the dark with the appropriate IRDye 800–conjugated secondary antibodies (1:5,000, LI-COR Bioscience, Inc.), the signals were detected using the Odyssey Imaging System (LI-COR Bioscience, Inc.).

For immunoblotting of cytosolic cytochrome c, cells were harvested, washed, and resuspended in 0.2 mL lysis buffer [250 mmol/L sucrose, 20 mmol/L HEPES (pH 7.5), 0.1 mmol/L EDTA, 0.1% fatty acid–free bovine serum albumin with proteinase inhibitor cocktail freshly added]. Cell suspensions were gently homogenized with a Dounce Homogenizer, and the homogenates were centrifuged (2,000 rpm, 15 min, 4°C). Supernatants were then centrifuged (12,000 rpm, 30 min, 4°C), after which the supernatants were used for identification of cytosolic cytochrome c by immunoblotting.

To visualize the mitochondrial fusion, the live A549 and HT-29 cells infected without or with Ad5-hHSG at 25, 50, 100 MOI were stained with MitoTracker Red CMXRos (150 nmol/L; Molecular Probes) and observed by confocal laser scanning microscopy (Leica TCS SP2, Leica Microsystems AG). For electron microscopy, the indicated cells were initially fixed in 0.01 mol/L sodium phosphate buffer containing 2.5% glutaraldehyde (pH 7.4). Next, cells were fixed, dehydrated, embedded, and sliced as described by Wang et al. (11). Ultrathin sections were stained with uranyl acetate and lead citrate and examined with a JEM-1230 transmission electron microscope (JEOL).

A549 and HT-29 cells were seeded in 60-mm dishes in DMEM plus 10% fetal bovine serum. After 12 h, medium was changed in DMEM plus 5% fetal bovine serum without (mock) or with Ad5-Null or Ad5-hHSG. Cells were collected at 24, 48, and 72 h and first stained with FITC-conjugated Annexin V for 30 min at room temperature and then stained with propidium iodide before 1 min analyzed by FACScan. Annexin V and propidium iodide were added according to the manufacturer's recommendations (BioVision). Detection and quantification of apoptotic cells was obtained by flow cytometry analysis (Becton Dickinson).

All animal experiments were done in accordance with the guidelines of the Institutional Animal Care and Use Committee. Female nu/nu nude mice, 4 weeks of age, were obtained from the Institute of Laboratory Animal Science and kept in a dedicated animal facility with five mice per cage. For ex vivo experiment, adenovirus-infected cancer cells were injected into the armpit of BALB/c nude mice in a total volume of 100 μL (5.0 × 10⁶ cells). When the tumors reached 4 mm in diameter, the tumor sizes in two perpendicular diameters were measured with precision calipers everyday. For in vivo experiment, 5 × 10⁶ cells were inoculated s.c. into the armpit of BALB/c nude mice. When the tumors reached 5 mm in diameter, 1 × 10⁹ plaque-forming units of Ad5-Null or Ad5-hHSG diluted in 50 μL adenovirus stock solution [4% sucrose, 10 mmol/L Tris (pH 8.0), 2 mmol/L MgCl₂] or adenovirus stock solution alone were intratumorally injected thrice (days 5, 10, and 15 for A549; days 5, 8, and 12 for HT-29). Tumor volume was calculated by using the standard formula: a × b² × 0.52 (a is the longest diameter; b is the shortest diameter). Animals showing severe distress or with tumors that exceeded 15 mm in two perpendicular diameters or 20 mm in one diameter were killed.

Cells were infected with Ad5-Null and Ad5-hHSG at 100 MOI when adhered. After infection for 12 h, cells were added with VP16 (etoposide, 1 mg/L; Sigma) or CHX (cycloheximide, 1 mg/L; Sigma) for chemotherapy or irradiated with different doses of γ-radiation (0, 4, 8, 12 Gy for A549 and 0, 2, 4, 6 Gy for HT-29) at room temperature for radiotherapy. After 48 h, treated cells were collected and stained with FITC-conjugated Annexin V and propidium iodide to analyze apoptosis as described above.

DiOC₆(3) is used to evaluate whether Ad5-hHSG–induced apoptosis is associated with the disruption of ΔΨm. After 48-h infection without or with adenoviruses at 100 MOI, cells were harvested and incubated with PBS containing 40 μmol/L DiOC₆(3) for 5 min at 37°C. Subsequently, analysis was done with FACSCalibur flow cytometer.

DEVDase activity analysis is a quantitative method of detecting caspase-3 as described (12). Samples were prepared in triplicate.

Results were expressed as mean ± SD. One-way ANOVA was used for statistical comparison among three groups, and post-hoc test (LSD) was used for statistical comparison between two groups. The differences between groups were considered to be statistically significant when the P value was <0.05.

Firstly, we constructed an adenovirus expression vector of hHSG and examined the role of Ad5-hHSG in cell survival. Cancer cell lines A549, HeLa, HT-29, MCF-7, and normal diploid cell lines BJ, L02, and HEK293 were infected in the absence or presence of 100 MOI of Ad5-Null and Ad5-hHSG. Using cell viability assay, we analyzed the antitumor effect of hHSG at 24, 48, and 72 h, respectively. As time elapsed, Ad5-hHSG overexpression significantly decreased the number of viable cells in all four cancer cell lines compared with diploid cell lines BJ, L02, and 293 (Fig. 1A). We further measured endogenous hHSG mRNA expression in these seven cell lines and found that all the cell lines expressed different levels of hHSG (Fig. 1B). These data indicate that overexpressing hHSG has a wide antitumor effect in vitro, whereas it has no effect on diploid cells.

To further investigate the antitumor activity of hHSG, we chose two epithelial cancer cell lines, A549 and HT-29, to examine. A549 and HT-29 cells were treated without or with Ad5-hHSG at 25, 50, and 100 MOI for 48 h. Treatment of A549 and HT-29 cells with Ad5-hHSG resulted in a significant dose-dependent overexpression of hHSG as assessed by Western blot (Fig. 2A-a,c). A549 and HT-29 cells were also treated with 100 MOI Ad5-hHSG for 24, 48, and 72 h. Treatment of A549 and HT-29 cells with Ad5-hHSG resulted in a significant time-dependent overexpression of hHSG (Fig. 2A-b,d).

To detect the mitochondrial morphologic change after overexpressing hHSG, confocal and electron microscopy were done. After 48-h infection with 25, 50, or 100 MOI of Ad5-hHSG, mitochondrial morphology was dramatically altered in cells infected with all three MOIs compared with cells treated with Ad5-Null or mock and ranged from conversion of the normal dispersed distribution of punctate mitochondria (Fig. 2B-a,b,f,g) to more reticular structures (Fig. 2B-c,d,h) and to extensive perinuclear clustering (Fig. 2B-e,i,j) by confocal microscopy. This result shows that hHSG overexpression causes mitochondrial fusion and morphologic changes in a dose-dependent manner. In electron microscopy, mitochondria shifted from being uniformly spread (Fig. 2C-a,b,f,g) into forming a complex tubular network (Fig. 2C-d,e,h), even a mass of swollen mitochondria clusters around nucleus at 100 MOI of Ad5-hHSG (Fig. 2C-c,i,j).

To determine whether the Ad5-hHSG–mediated decrease in viable cells occurred via apoptosis, A549 and HT-29 cells were analyzed for apoptotic changes by fluorescence-activated cell sorting analysis. After 48 h infection with mock, Ad5-Null, Ad5-hHSG, and Ad5-P53 at 100 MOI, the percentage of apoptotic cells in response to Ad5-hHSG or Ad5-P53 increased by 1.33× and 0.57× over Ad5-Null in A549 cells and 2.14× and 1.74× in HT-29 cells, respectively (Fig. 3A). Ad5-hHSG–induced apoptosis of both cell types occurred in time-dependent and dose-dependent manner. At 24 h, hHSG-induced cancer cell apoptosis was not significantly greater than that of the Ad5-Null–treated controls. At 48 h, Ad5-hHSG or Ad5-P53 treatment resulted in a significant dose-dependent increase in the number of apoptotic cells at both the early and late stages of apoptosis. The percentage increase after Ad5-hHSG infection ranged from 8.21% to 36.68% in A549 cells and from 5.24% to 26.71% in HT-29 cells, which was higher than that induced by Ad5-P53 infection, which ranged from 6.00% to 25.07% in A549 cells and from 5.94% to 15.67% in HT-29 cells, depending on viral titer. No changes were observed in cancer cells treated with Ad5-Null or mock. By 72 h, Ad5-hHSG still induced higher amounts of apoptosis than Ad5-P53 (Fig. 3B).

To further examine the possible inhibitory effect of hHSG on cell cycle progression, we infected A549 and HT-29 cells for 48 h with 25, 50, and 100 MOI Ad5-hHSG and analyzed cell cycle stages by fluorescence-activated cell sorting. The number of A549 cells in G₀-G₁ phase increased significantly from 7.53% at 25 MOI to 14.06% at 50 MOI and to 17.68% at 100 MOI compared with Ad5-Null–treated cells. In infected HT-29 cells, the cell cycle was arrested in G₂-M, and the percentage increased from 4.51% at 25 MOI to 5.93% at 50 MOI and to 26.66% at 100 MOI compared with Ad5-Null–treated cells. No obvious changes were observed between Ad5-Null–treated and adenovirus stock solution–treated cells (Supplementary Fig. S3).³

We next sought to determine whether exogenous hHSG affected tumorigenicity both ex vivo and in vivo. Cells were treated with adenovirus stock solution (mock) or with 100 MOI of Ad5-Null and Ad5-hHSG and injected into nude mice 12 h later. Ad5-hHSG–treated tumor cells almost completely suppressed tumor growth ex vivo, whereas Ad5-Null–treated and adenovirus stock solution–treated cells resulted in tumor growth (P < 0.05; Table 1A and B).Our results showed that both A549 and HT-29 cells' tumorigenicity was completely inhibited by Ad5-hHSG, indicating that hHSG protein may have therapeutic efficacy. The therapeutic potential of Ad5-hHSG was further examined using an in vivo tumor model. Nude mice were inoculated with tumor cells and injected with adenovirus stock solution, Ad5-Null, or Ad5-hHSG after tumorigenesis. Tumor growth was significantly inhibited after Ad5-hHSG treatment from 9th day in A549 and from 14th day in HT-29 compared with Ad5-Null treatment (P < 0.05; Table 1C and D).

To examine the effect of Ad5-hHSG on chemotheraphy and radiotherapy, chemosensitization and radiosensitization experiments were done. We began by evaluating the antitumor effect of Ad5-hHSG on A549 and HT-29 cells when used in combination with chemotherapy in vitro and measuring apoptosis by fluorescence-activated cell sorting. The chemotherapeutic agents VP16 and CHX were used in the combination experiments. Ad5-hHSG and VP16 combination therapy resulted in 47.19 ± 1.76% apoptosis in A549 cells, which was higher than VP16 (14.37 ± 2.14%) or Ad5-hHSG alone (24.18 ± 2.83%). Ad5-P53 and VP16 combination therapy only reached 29.72 ± 3.87%, although this was higher than single treatment with Ad5-P53 (24.53 ± 2.06%) or VP16 (14.37 ± 2.14%). Ad5-P53 also had a synergistic effect on CHX sensitization (23.76 ± 2.81%) in A549 cells; however, Ad5-hHSG had an even stronger effect (42.23 ± 2.91%; Fig. 4A). Ad5-hHSG also significantly increased the VP16 (40.15 ± 4.21%) and CHX (32.93 ± 1.99%) sensitivity of HT-29 cells and had a greater effect than Ad5-P53 and VP16 (27.41 ± 5.56%) or CHX (21.15 ± 3.40%; Fig. 4B).

We also observed the combining effect of Ad5-hHSG and radiotherapy. Ad5-hHSG treatment of A549 cells only produced modest tumor cell apoptosis (21.31 ± 2.82%), whereas treatment with γ-radiation increased apoptosis from 17.81 ± 0.66% to 33.64 ± 1.95%. Combination therapy with Ad5-hHSG and radiation resulted in between 38.64 ± 1.93% and 48.95 ± 0.97% apoptosis, which was slightly lower than Ad5-P53 and radiation combination treatment (between 48.59 ± 2.07% and 57.55 ± 2.07%; Fig. 4C). In HT-29 cells, Ad5-hHSG and radiation treatment ranged from 38.67 ± 3.30% to 48.61 ± 1.28% and was higher than radiation and Ad5-P53 treatment, which ranged from 16.65 ± 1.06% to 24.73 ± 5.36% (Fig. 4D). Because the synergic antitumor activity of Ad5-hHSG is independent of the state of endogenous hHSG, this makes it a better agent than Ad5-P53.

To evaluate whether the apoptotic effect of hHSG is associated with the mitochondrial apoptotic pathway, ΔΨm was measured. Loss of ΔΨm was visualized as a reduction in the fluorescence signal in the FL1 channel. Infection of A549 and HT-29 cells with 100 MOI of Ad5-hHSG for 48 h reduced ΔΨm. The percentage of negative fluorescence induced by Ad5-hHSG in A549 cells was 26.55% compared with 10.40% and 6.28% induced by mock and Ad5-Null, respectively. Ad5-hHSG induced 13.00% in HT-29 cells compared with 6.22% and 5.08% induced by mock and Ad5-Null, respectively (Fig. 5A). Involvement of the mitochondrial apoptotic pathway was confirmed by the release of cytochrome c from mitochondria to cytosol after infection (Fig. 5B). We also analyzed caspase-3 activity in infected cells using DEVDase activity analysis (13) and Western blot. At 48 h postinfection with 100 MOI, DEVD cleavage in lysates from hHSG-infected A549 cells was at least four times higher than from mock or Ad5-Null–infected cell lysates. hHSG infection of HT-29 cells resulted in 21.8× higher DEVD cleavage (Fig. 5C). PARP, as a representative substrate of caspase-3, was cleaved in Ad5-hHSG–infected cells (Fig. 5D).

In this study, we firstly confirmed the antitumor activity of Ad5-hHSG in a wide range of cancer cell lines. Using adenovirus-expressed hHSG, we found that it significantly affected the survival of different epithelial cancer cell lines. However, it was not efficient in some diploid cell lines, suggesting that its effect on cell survival has a tendency toward cancer cells. It still remains unknown why normal cells are insensitive to Ad5-hHSG. To investigate the antitumor activity of Ad5-hHSG and its mechanism, we selected two cancer cell lines, A549 and HT-29, and did a series of experiments. Ad5-hHSG could induce mitochondrial perinuclear clustering, trigger apoptosis, arrest cell cycle, decrease the ΔΨm, and subsequently release cytochrome c in both cancer cell lines. In an animal model, the tumorigenicity of A549 and HT-29 was significantly inhibited by Ad5-hHSG treatment. The antitumor activity of Ad5-hHSG is independent of the state of endogenous hHSG, making it even more effective than P53. In addition, Ad5-hHSG significantly increased the sensitivity of cancer cell lines to chemotherapeutic agents and γ-radiation.

rHSG (rMfn2) is a powerful regulator of cell proliferation both in vitro and in vivo (1). Overexpression of rHSG with the adenovirus vector Ad5-rHSG inhibited mitogenic stimuli–mediated proliferation of cultured Wistar-Kyoto vascular smooth muscle cells and blocked balloon injury induced by neointimal vascular smooth muscle cell proliferation and restenosis in vivo. However, when Mfn2 was overexpressed using eukaryotic expression vectors, Mfn2 and its variant, Mfn2RasG12V (“activated Mfn2”), inhibited Bax activation, release of cytochrome c, and cell death (14–16). These results contrast with our current findings. Why would the same gene have a different effect in different laboratories? One possibility is that adenovirus expression vectors allow for higher expression level of hHSG than plasmid expression vectors. Before this study, we compared hHSG expression using plasmid and adenovirus vectors and found that expression levels were much lower with the plasmid vector and there was no apoptosis although mitochondrial fusion was also observed (Supplementary Fig. S5).³ When the eukaryotic expression plasmid containing hHSG was over highly expressed, we could detect cell growth inhibition and few cancer cells undergoing apoptosis (17). With the adenovirus vector Ad5-hHSG, mitochondrial perinuclear clustering appeared at 25 MOI, whereas swollen mitochondrial clusters and extensive cancer cell apoptosis were observed as the dose increased to 100 MOI. Our results suggest that Ad5-hHSG–mediated overexpression significantly elevated cancer cell apoptosis.

The antitumor mechanism of hHSG remains unclear. We first supposed that hHSG might be involved in cancer based on its location on chromosome 1p36.3. Cytogenetic analysis of various types of human malignancies, ranging from solid tumors to leukemias and myeloproliferative disorders, have shown nonrandom involvement of abnormalities on chromosome 1p36 (18, 19). In the past few decades, many tumor suppressor genes have been found in this region (20, 21). In our preliminary study, we detected endogenous hHSG in both normal and cancerous tissue by immunohistochemistry and in situ hybridization; however, the positive rate of hHSG was lower in cancer (51.51%) than normal tissue (68.96%; data not shown). Down-regulated expression or loss-of-function of hHSG in cancers might have a role in the development of cancers.

In addition, as a mitochondrial dynamic protein, hHSG overexpression has an effect on mitochondrial morphology and functions. Our findings show that hHSG induced cancer cell apoptosis, along with ΔΨm reduction and release of cytochrome c, assuming that Ad5-hHSG–induced apoptosis is associated with Bax. Bax is a member of the Bcl-2 family of proteins known to regulate mitochondria-dependent programmed cell death (22, 23). Early in apoptosis, Bax translocates from the cytosol to the mitochondrial membrane. Then Bax leaves the mitochondrial membranes and coalesces into large clusters containing thousands of Bax molecules that remain adjacent to mitochondria (24). Bax proapoptotic mechanism is immediately subsequent to mitochondrial translocation. It has been proposed that Bax interacts with the coiled-coil domain of hHSG (Mfn2) to induce mitochondrial fusion (8). However, in our experiments, after we silenced endogenous Bax expression using small interfering RNA, the apoptotic inhibitory rate of siBax group was only approximately 20% compared with nonsilencing group (Supplementary Fig. S6A,B).³ Our results suggested that siBax could partly inhibit Ad5-hHSG-induced apoptosis. Also, we found that silencing endogenous Bax prevented the release of cytochrome c to cytosol, but it did not affect the cleavage of PARP (which is a substrate of caspase-3). As is known to us, caspase-3, a key factor in apoptosis execution, participates in mitochondrial and nonmitochondrial apoptotic pathway. These results further proved that Ad5-hHSG–induced apoptosis was via both Bax-dependent mitochondrial apoptotic pathway and Bax-independent apoptotic pathway. Another possibility is that Ad5-hHSG–induced mitochondrial excessive fusion directly leads to the irreversible pathologic changes of mitochondrial morphology and induces apoptosis.

At present, multiple therapies are used to treat cancers. Intratumoral injection of Ad-P53 (INGN201) in combination with radiation therapy is well tolerated and shows evidence of primary injected tumor regression (25). Exogenous overexpression of P53 in combination with DNA-damaging agents may increase susceptibility to apoptosis specifically by tumor cells (26). Our previous study suggests that hHSG inhibits tumor cell proliferation and increases their sensitivity to chemotherapy (17). In this study, we are excited to find that Ad5-hHSG increased the radiosensitization and chemosensitization of A549 and HT-29 cells even above Ad5-P53. Although additional in vivo experiments are needed, the knowledge that Ad5-hHSG sensitizes A549 and HT-29 cells to VP16, CHX, and radiation may provide a foundation for new therapeutic approaches.

In summary, howbeit hHSG is a gene participating in mitochondrial fusion and overexpression of Ad5-hHSG has a strong antitumor activity in vitro and in vivo and sensitizes cells to chemotherapy and radiation therapy more effectively than Ad5-P53. Our present data indicate that Ad5-hHSG is a potential gene therapeutic drug.

We thank Shigehisa Hirose for the anti-HSG rabbit polyclonal antibody, Prof. Zhiqian Zhang and Lan Yuan for help with confocal laser scanning microscope.

This manuscript was proofread by an English speaking professional with science background at Elixigen Corporation.