Prostate cancer is the most commonly diagnosed malignant tumor.1 In the USA, prostate cancer is the second leading cause of cancer-related death in men. Prostate cancer develops slowly and tumor growth is androgen dependent. Following androgen ablation or castration, prostate cancer regresses but later adopts an androgen-refractory phenotype. Androgen-refractory prostate cancer is more resistant to a number of therapeutic modalities, including chemotherapy and radiotherapy, and is ultimately lethal. Therefore, novel therapeutic approaches are required to treat androgen-refractory prostate cancer.2, 3 Gene therapy is one of the most promising strategies in the fight against cancer. For the treatment of localized prostate cancer, a recombinant adenovirus vector containing a herpes simplex virus Type-1 thymidine kinase gene has been directly injected into the prostate followed by the administration of gancyclovir.4 Liposomes containing the IL-2 gene have also been used.5 Generally, these trials have produced transient responses with minimal toxicity. For the treatment of metastatic prostate cancer, immune gene therapy has been investigated, including the use of autologous tumor cells expressing granulocyte–monocyte-colony stimulating factor6, 7 and vaccinia virus expressing granulocyte–monocyte-colony stimulating factor or prostate-specific antigen.8–10 To enhance therapeutic efficacy, additional treatments such as radiotherapy have been combined with gene therapy, and a number of oncolytic viruses have been developed by viral gene engineering.11–13 To achieve tumor-selective killing, tumor-specific promoters are being used to drive virus replication in cancer cells. For the treatment of replication-competent adenovirus driven by prostate-specific antigen promoter has been locally or intravenously administered to patients.14 In some cases, oncolytic viruses are armed with therapeutic genes, such as cytosine deaminase and herpes simplex Type-1 thymidine kinase. These oncolytic viruses appear to be more effective for cancer killing than conventional gene therapy using replication-defective vectors, but also carry a number of risks. Therefore, trials involving oncolytic viruses are being cautiously undertaken with dose-escalation.15, 16
Natural mutant viruses, such as E1B-deficient adenovirus and accessory gene-deleted herpes virus, also selectively replicate in cancer cells.16, 17 Some viruses, such as vaccinia virus and reovirus, have been used for oncolysis following systemic administration.18, 19 Paramyxoviruses, including Newcastle disease virus and the Edmonston strain of measles virus, are also being used for cancer treatment in humans.20, 21 Sendai virus, a paramyxoviridae, is famous for its robust fusion activity and ability to induce Type I interferon (IFN).22, 23 Recombinant Sendai virus vector has been constructed to achieve enhanced expression of a transgene within the cytoplasm.24 Inactivated Sendai virus particles [i.e. hemagglutinating virus of Japan envelope (HVJ-E)] have also been used as an effective delivery system for genes, proteins and siRNA, both in vitro and in vivo.25, 26 Furthermore, HVJ-E has been used in other anti-cancer treatment strategies.26 We have previously demonstrated that HVJ-E can eradicate mouse tumors, such as colon cancer (CT26) and renal cancer (Renca), by activating dendritic cells (DC).27, 28 Following HVJ-E treatment, DCs secrete various cytokines and chemokines, such as Type I IFN, IL-6 and CXCL10, resulting in the activation of cytotoxic T lymphocytes (CTL) and natural killer (NK) cells.27, 28 IFN-α and -β have generated a lot of interest as anti-viral and anti-tumor cytokines.29 However, Type I IFN was not produced in mouse tumor cells following HVJ-E treatment, and HVJ-E failed to directly kill mouse tumor cells in the absence of DCs.28
In the present study, we demonstrated that hormone-resistant human prostate cancer cells were driven into suicide by HVJ-E-induced Type I IFN production and that PC3 tumors in SCID mice were efficiently eradicated by HVJ-E treatment.
CTL, cytotoxic T lymphocytes; DAPI, 4′,6-diamino-2-phenylindole; DC, dendritic cell; HVJ-E, hemagglutinating virus of Japan envelope; IFN, interferon; IFNRAb, IFN-α/β receptor antibody; JAK, Janus kinases; MOI, multiplicity of infection; NK, natural killer; RIG-I, retinoic acid-inducible gene-I; STAT, signal transducers and activators of transcription.
Material and methods
Preparation of HVJ-E
HVJ (VR-105 parainfluenza 1 Sendai/52, Z strain) was purified from the chorioallantoic fluid from 10 to 14 day-old chick eggs, after which it was purified by centrifugation and inactivated by UV irradiation (99 mJ/cm2), as previously described.25 Inactivated virus cannot replicate, but its capacity for viral fusion remains intact.25, 28
Reagents and antibodies
CellTiter 96® Aqueous One Solution Cell Proliferation Assay was purchased from Promega (Madison, WI). Anti-human cleaved caspase 3 antibody (PC-020) was purchased from TREVIGEN (Gaithersburg, MD). Anti-human caspase 8 antibody (H-134), anti-human caspase 9 antibody (H-83), and anti-human p-STAT1α (Tyr701) antibody (sc-7988), were purchased from Santa Cruz (Santa Cruz, CA). Anti-human caspase 8 antibody (AM46T) was also purchased from Calbiochem (San Diego, CA). Anti-human β-actin antibody (IMG-5142A) was purchased from IMAGENEX (San Diego, CA). The in situ apoptosis detection kit was purchased from Takara Bio (Shiga, Japan). The Annexin V-FITC apoptosis detection kit was purchased from BD Bioscience (San Diego, CA). Anti-human IFN-α/β receptor chain 2 (CD118) antibody (IFNRAb) was purchased from PBL Biomedical Laboratories (Piscataway, NJ). JAK inhibitor I was purchased from Calbiochem (San Diego, CA). Anti-mouse asialo GM1 antibody (CL8955) was purchased from Cedarlane Laboratories Limited (Ontario, Canada).
Cell culture and mice
Hormone-resistant human prostate cancer cell PC3 and DU145, and hormone-sensitive human prostate cancer cell, LNCap clone FGC were purchased from American Type Culture Collection (Rockville, MD). Human normal prostatic epithelial cell, PNT2 was purchased from the European Collection of Animal Cell Cultures (Porton Down, UK). PC3 cells were maintained in Dulbecco's modified Eagle F12 medium (Nakarai Tesque, Kyoto, Japan), and DU145, LNCap and PNT2 cells were maintained in RPMI 1640 medium (Nakarai Tesque, Kyoto, Japan), with 10% fetal bovine serum (FBS), 100 U/ml penicillin and 100μg/ml streptomycin. Cells were incubated at 37°C in a humidified atmosphere of 95% air and 5% CO2. Five to six-week old male C.B-17/IcrCrj-SCID mice were purchased from Charles River (Yokohama, Japan) and maintained in a temperature-controlled, pathogen-free room. All animals were handled according to the approved protocols and guidelines of the Animal Committee of Osaka University.
Pretreatment of JAK inhibitor and IFN-receptor antibody
For blockade of JAK/STAT (Janus kinases/Signal Transducers and Activators of Transcription) signaling activated by Type I IFN, cells were pre-incubated in medium containing JAK inhibitor I (1 μM) (Calbiochem, San Diego, CA) to inhibit JAK 1, 2 and 3 or IFNRAb (20 μg/ml) for 3 hr before HVJ-E treatment. The concentration of the reagents was determined by the previous reports28 and manufacturer's protocol.
Cell viability assay
Cancer cells were seeded in 96-well plates (1 × 104 cells/well). Twenty-four hours later, the cells were treated with HVJ-E [multiplicity of infection (MOI):10–104]. Twenty-four hours after HVJ-E treatment, cell survival was assessed by measuring absorbance at 490 nm (A490) after adding 20 μl of CellTiter 96® Aqueous One Solution Reagent (Promega).
Microscopic observation of HVJ-E treated cells
Cells were seeded in 6-well plates (2 × 105 cells/well). The next day, cells were treated with HVJ-E (MOI: 104). Then, 24 hr after HVJ-E treatment, cells were observed by microscopy.
Interaction of PKH26-labeled HVJ-E with cells
HVJ-E (3 × 1010 particles) were suspended in 1 ml of Diluent C buffer and incubated with 1 ml of 4 mM PKH26 dye in Diluent C buffer solution at room temperature for 5 min according to manufacturer's instructions (PKH26 Red Fluorescent Cell Linker Kit; Sigma, St. Louis, MO). Labeling was stopped with 2 ml FBS. Then, labeled HVJ-E was washed 3 times with PBS and resuspended in 3 ml of PBS (1 × 1010 particles/ml). The day before HVJ-E treatment, cells were plated onto polyethyleneimine (PEI)-coated cover glasses in 6-well plates (2 × 105 cells/well). Cells were incubated with PKH26-labeled HVJ-E (MOI: 104) at 37°C for 1 hr, washed with PBS twice and fixed with 4% paraformaldehyde. Finally, cells were stained with 4′,6-diamino-2-phenylindole (DAPI) and observed by a confocal microscope (Radiance2100; Bio-Rad Japan, Tokyo, Japan).
Cancer cells were seeded at 1 × 105 cells/well on PEI-coated cover glasses in 6-well plates. The following day, the cells were treated with HVJ-E (MOI: 104). Twenty-four hours after HVJ-E treatment, apoptotic cells were observed. Cells were washed twice with PBS, and fixed with 4% paraformaldehyde for 15 min at 4°C. The terminal deoxynucleotide transferase-mediated dUTP nick-end labeling (TUNEL) assay was carried out according to the protocol of an in situ apoptosis detection kit (Takara Bio).
Annexin V staining
An Annexin V-FITC apoptosis detection kit was used to determine the percentage of apoptotic cells observed after HVJ-E treatment with or without pretreatment with a JAK inhibitor. Cancer cells were seeded at 1 × 105 cells/well in 6-well plates. The following day, the cells were treated with HVJ-E (MOI: 104) with or without pre-incubation with a JAK inhibitor (1 μM) for 3 hr. Twenty-four hours after adding HVJ-E, the treated cells were washed twice with PBS and re-suspended in labeling solution. Five microliters of annexin V and 2 μl of propydium iodide from BD PharMingen (San Diego, CA) were added into the re-suspended solution and incubated in the dark for 15 min at room temperature. Following this, the stained cells were analyzed with a FACScan flow cytometer (Becton Dickinson) using Cell Quest software.
Western blot analysis
The harvested human cancer cells were lysed in lysis buffer as previously described.30 After adding the same amount of sample buffer, the cell lysates were boiled for 10 min. Ten micrograms of protein from each cell lysate was electrophoresed on a 4–20% sodium dodecyl sulfate polyacrylamide gel. After transfer onto a polyvinylidene fluoride membrane, the membrane was blocked with 5% skim milk, after which the blots were incubated overnight at 4°C with primary antibodies at a 0.1% antibody concentration in 5% skim milk. The membranes were washed and labeled with horseradish peroxidase conjugated anti-mouse, anti-rabbit (Amersham, Buckinghamshire, UK) or anti-goat antibodies (Jackson Immuno-Laboratories, West Grove, PA) at room temperature for approximately 1 hr. Detection by chemiluminescence was performed following the standard protocol outlined in the ECL user's guide (Amersham, Buckinghamshire, UK).
PC3 cells (1 × 106 cells) were seeded in a 10 cm dish. The following day, the cells were treated with HVJ-E (MOI: 104). Twelve hours after adding HVJ-E, RNA was isolated using an RNeasy Mini Kit (Qiagen, Tokyo, Japan) according to the manufacturer's instructions. Following the isolation of RNA, all subsequent technical procedures, including quality control of RNA, labeling, hybridization and scanning of the arrays, were performed at Bio Matrix Research, (Chiba, Japan). Complementary DNA and biotin-labeled cRNA were synthesized according to protocols for Affymetrix array analysis. Biotin-labeled cRNAs (15 μg) were hybridized to the GeneChip Human Genome U133 Plus 2.0 Array (Affymetrix, Santa Clara, CA). To determine the average differences for each probe set, changes were calculated by a global normalization method using Affymetrix GCOS (GeneChipOperating Software) software.
PC3 cells were seeded at 5 × 104 cells/well in 96-well plates. The following day, HVJ-E (MOI: 102–104) was added and the culture continued for another 24 hr. After this, IFN-α and -β were measured within the supernatant by ELISA using commercially available reagents (PBL Biomedical Laboratories, Piscataway, NJ).
The analysis of acidic glycosphingolipids from prostate cancer cells
PC3, DU145, LNcap and PNT2 (2 × 105 cells) were extracted with 1,200 μl of chloroform/methanol (2:1, v/v), followed by 800μl of chloroform/methanol/water (1:2:0.8, v/v/v). Both extracts were loaded onto a DEAE-Sephadex A25 column and acidic glycosphingolipids (GSLs) eluted with 200 mM ammonium acetate in methanol. Acidic GSLs were desalted by gel filtration on a HW-40C column (Tosoh) (LMS, Tokyo, Japan) equilibrated with chloroform/methanol/water (5:5:1, v/v/v). Acidic pyridylaminated (PA)-oligosaccharides were prepared and separated on a Shimazu LC-20A high performance liquid chromatography (HPLC) system equipped with a Waters 2475 fluorescence detector. Size fractionation HPLC was performed on a TSK-gel® Amide-80 column (0.2× 25 cm2, Tosoh) at 40°C at a flow rate of 0.2 ml/min using 2 solvents, A and B. Solvent A was acetonitrile/0.5 M acetic acid containing 10% acetonitrile, adjusted to pH 7.3 with triethylamine (75:15, v/v). Solvent B was acetonitrile/0.5 M acetic acid containing 10% acetonitrile, adjusted to pH 7.3 with triethylamine (40:50, v/v). The column was equilibrated with solvent A. After injection of sample, the proportion of solvent B was programmed to increase from 0 to 100% in 100 min. The PA-oligosaccharides were detected by fluorescence with an excitation wavelength of 310 nm and an emission wavelength of 380 nm. The molecular size of each PA-oligosaccharide is given in glucose units (Gu) based on the elution times of PA-isomaltooligosaccharides. Reversed phase HPLC was performed on a TSK-gel® ODS-80Ts column (0.2 × 15 cm2, Tosoh) at 30°C at a flow rate of 0.2 ml/min using 2 solvents, C and D. Solvent C was 50 mM acetic acid, adjusted to pH 6.0 with triethylamine. Solvent D was 50 mM acetic acid containing 20% acetonitrile, adjusted to pH 6.0 with triethylamine. The column was equilibrated with solvent C. After injection of sample, the proportion of solvent D was programmed to increase from 0 to 18% in 54 min. The PA-oligosaccharides were detected by excitation at 315 nm and emission at 400 nm. The retention time of each PA- oligosaccharide is given in glucose units based on the elution times of PA-isomaltooligosaccharides. Thus, a given compound on these 2 columns provides a unique set of Gu (amide) and Gu (ODS) values, which correspond to coordinates of the 2-dimensional map (2-D map). PA-oligosaccharides were analyzed by LC/ESI MS/MS as described previously.31 Standard PA-oligosaccharides, PA-GM1, PA-GD1a were purchased from Takara Bio, and PA-LST-a and PA-SPG (sialylparagloboside) were obtained from our previous study.31
Tumor growth in vivo
Viable PC3 cells (2 × 106 cells) were re-suspended in 100 μl of PBS and intradermally injected into the backs of the mice. When each tumor had grown to ∼4–6 mm in diameter, the mice were treated with intra-tumor injections of HVJ-E (1.5 × 1010 particles in a total volume of 100 μl) or 100 μl of PBS on days 10, 13 and 16. Tumor volume was measured in a blinded manner with slide calipers using the following formula: tumor volume (mm3) = length × (width)2/2. To deplete NK cells in vivo, 40 μl of an anti-asialo GM1 antibody was simultaneously co-injected into each tumor with HVJ-E on days 10, 13 and 16.
Data are expressed as the mean ± S.E.M. Statistical comparisons between groups were performed with the student's t-test. A value of p < 0.05 was considered significant.
When we treated the hormone-resistant human prostate cancer cells, PC3 and DU145, with various amounts of HVJ-E, the viability of both cell lines was significantly suppressed by HVJ-E in a dose-dependent manner (Fig. 1a). However, growth inhibition of hormone-sensitive LNCap cell and normal prostate epithelium was not observed. Microscopic observation showed that HVJ-E extensively induced cell fusion (arrows) in both PC3 cells and DU145 cells while cell fusion hardly occurred in LNCap and PNT2 cells (Fig. 1b). We tested the ability of PKH26-labeled HVJ-E to bind to prostate cancer cells and normal prostate epithelium. As shown in Figure 2a, red fluorescence-labeled HVJ-E was abundantly incorporated into PC3 and DU145, but not into LNCap and PNT2. Thus, fusion of HVJ-E with the cell membrane predominantly occurred in hormone-resistant prostate tumor cells and appeared to suppress the growth of these prostate cancer cells.
Then, we determined the quantity of gangliosides for Sendai virus receptors in prostate cancer cells and normal prostate epithelium using HPLC (Fig. 2b). From comparison of the positions on the map to the positions of standard acidic PA-oligosaccharides, Peak 1 of 4 cells is predicted to be GM1. Peak 2 from the 4 different cells consisted of SPG, GD1a and LST-a. Because both GD1a and SPG are receptors for HVJ,32 Peaks 2 were further purified by reversed phase HPLC. Peaks 2 from PC3, DU145, LNcap and PNT2 were separated into 2, 3, 1 and 2 major components, respectively (Fig. 2c). Purified acidic PA-oligosaccharides were subjected to LC/ESI MS/MS and the total amount of GD1a and SPG was compared among these cell lines. The ratio of total Sendai virus receptor gangliosides in PC3, DU145, LNCap and PNT2 was 68.5, 49.4, 1 and 26.4, respectively (Fig. 2c). Thus, hormone-refractory prostate cancer cells produced higher amount of GD1a and SPG than LNCap and PNT2.
Approximately, 3.8% of apoptotic cells (6/156 cells) were detected 24 hr after HVJ-E treatment by TUNEL staining in PC3 cells while no apoptotic cells (0/189 cells) were seen in the control (Fig. 3a). FACS analysis showed that approximately 10% of the PC3 cells were Annexin V-positive in the control group. However, in the HVJ-E treatment group, Annexin V-positive cells exceeded 20% (Fig. 3b). A significant difference in the number of apoptotic PC3 cells was observed after HVJ-E treatment (p < 0.05).
Following this, we examined caspase activity by western blot analysis (Fig. 3c). Increased expression of caspase 3 and 8 was detected in PC3 cells after HVJ-E treatment, whereas the expression of caspase 9 remained unchanged by HVJ-E.
Next, we investigated the gene expression profile of PC3 cells 12 hr after HVJ-E treatment. Microarray analysis revealed 17 IFN-induced genes (underlined)33–38 among the 20 most up-regulated genes (Table I). This result suggests that the secretion of IFN by tumor cells following HVJ-E treatment might stimulate IFN-related genes.
Table I. Microarray Analysis of Genes Upregulated by HVJ-E Treatment in PC3 Cells
PC3 cells (1 × 106 cells) were incubated with or without HVJ-E (MOI: 104) for 12 hr. After this, RNA isolated from each sample was subjected to microarray analysis. The 20 most up-regulated genes in HVJ-E treated PC3 cells, compared with untreated controls, are shown. Seventeen IFN-induced genes (all underlined) (31 – 35) were included among the 20 most up-regulated genes.
Interferon-induced protein with tetratricopeptide repeats 1
2′,5′-oligoadenylate synthetase 1
Myxovirus(influenza virus) resistance 1
Zinc finger CCCH-type
2′,5′-oligoadenylate synthetase 1
Hypothetical protein FLJ20035
Interferon induced with helicase C domain 1
DEAD (Asp-Glu-Ala-Asp) box polypeptide 58
Dehydrogenase/reductase (SDR family) member 2
Interferon-induced protein 44
Sterile alpha motif domain containing 9
Interferom-induced protein with tetratricopeptide repeats 3
Hect domain and RLD 5
2′-5′-oligoadenylate synthetase 2
Interferon–stimulated trascription factor 3
Interferon induced transmembrane protein 1
Interferon alpha-inducible protein 6
Hect domain and RLD 6
2′-5′-oligoadenylate synthetase 3
It is known that live HVJ induces the production of Type I IFN by retinoic acid-inducible gene-I (RIG-I)-mediated recognition of double-stranded viral RNA.39 We have previously reported that HVJ-E also induces Type I IFN secretion in murine DCs in a fusion-dependent manner,40 but not in murine cancer cells.27, 28 However, in the culture medium of PC3 cells, significant induction of IFN-α and -β was detected following HVJ-E treatment (Fig. 4a). Those IFNs were also secreted from DU145 cells, neither PNT2 nor LNCap cells (data not shown). RIG-I expression increased in both PC3 cells (Fig. 4b) and DU145 cells (data not shown) after HVJ-E treatment, and this induction was suppressed by antibody against the IFN receptor (Fig. 4b). The densitometrical measurement30 revealed that RIG-I expression was reduced to ∼51% by the IFNRAb compared with that in the absence of the antibody. The expression of RIG-I without HVJ-E treatment was hardly detected in PC3 cells by western blot as seen in the previous report.41 Thus, HVJ-E augments production of the cytoplasmic receptor of its RNA genome through the production of Type I IFN (Fig. 4b).
Type I IFN is known to induce apoptosis in some transformed cell lines. As shown in Figure 3c, caspase 3 and 8 were up-regulated in HVJ-E-treated PC3 cells. Caspase 8 transcription is known to be activated by JAK2/STAT1 signaling.42 Thus, we blocked the JAK/STAT signaling pathway using a JAK inhibitor. Following HVJ-E treatment, phosphorylated STAT1 (pSTAT1α) was greatly activated in PC3 cells. However, by adding a JAK inhibitor to PC3 cells before HVJ-E treatment, pSTAT1 was reduced to an undetectable level even after HVJ-E treatment (Fig. 5a). Caspase 8 was also down-regulated in the presence of the inhibitor. When an antibody against the IFN receptor was added to PC3 cells 3 hr before HVJ-E treatment, STAT1 phosphorylation and caspase 3 and 8 activation by HVJ-E were highly suppressed. Annexin V-positive cells increased with HVJ-E, but a significantly reduced number of apoptotic PC3 cells were observed following pre-treatment with the JAK inhibitor (Fig. 5b).
Next, we examined whether intra-tumor injection of HVJ-E inhibited tumor growth. PC3 cells were intradermally inoculated into the backs of SCID mice, after which each tumor mass was directly injected with HVJ-E or saline 3 times on days 10, 13 and 16 after inoculation. In the HVJ-E-treated group, the tumors were completely eradicated in all mice (n = 3) (Figs. 6a and 6b). This experiment was repeated 4 times with the same result (total n = 13). It is known that HVJ-E eradicates mouse renal cancers by activating NK cells.28 Then, to test the effect of NK cells on tumor regression by HVJ-E treatment, we co-injected an anti-asialo GM1 antibody with HVJ-E 3 times into each PC3 tumor mass, according to a previously reported protocol.28 Tumor regression by HVJ-E was significantly attenuated by the anti-asialo GM1 antibody on days 60 and 70 (*p < 0.05). However, significant suppression of tumor growth was still observed by HVJ-E treatment even in the presence of anti-asialo GM1 antibody compared with the mice treated with PBS (**p < 0.05) (Fig. 6c).
Here, we demonstrated the ability of HVJ-E to directly kill hormone-resistant human prostate cancer by inducing tumor cell apoptosis via IFN secretion from tumor cells.
Apoptosis of hormone-resistant prostate cancer cells occurred because of the secretion of IFN by cancer cells treated with HVJ-E, and not from normal prostate epithelium. It is known that components of the viral genome must enter the cytoplasm because the cytoplasmic receptor, RIG-I helicase, must recognize double-stranded RNA of the Sendai virus genome to transmit a signal to IRF-3, which activates IFN transcription.39, 41 Therefore, tumor cell apoptosis appears to depend on the preferential binding of HVJ-E to hormone-resistant prostate cancer cells. The receptors of HVJ are GD1a and SPG.32 Both gangliosides bearing N-acetylneuraminic acid attached on terminal galactose residues by α2-3 linkages that are recognized by HN protein of HVJ.32 These molecules are thought to be widely distributed among various cell lines.32 However, when the cell surface gangliosides within prostate epithelium was precisely examined, GD1a gangliosides were found to be the most prevalent gangliosides among prostate cancer cell lines, including PC3 and DU145, whereas GD1a gangliosides were barely detectable in both hormone-sensitive prostate cancer LNCap cells and normal prostate epithelium using thin-layer chromatography.43, 44 Furthermore, when we determined the quantity of gangliosides in prostate cancer cell lines using HPLC, we found that SPG, another receptor for HVJ, also increased in hormone- refractory prostate cancer cells as well as GD1a. The total amount of GD1a and SPG was much higher in PC3 and DU145 than LNCap and PNT2 (Fig. 2c). The preferential binding of HVJ-E to PC3 and DU145 in Figure 2a is consistent with the results of the cell surface ganglioside analysis in Figure 2c. GD1a is also present on the epithelium of prostate cancer cells at sites of metastasis.44 Thus, when HVJ-E is directly injected into prostate tissue, it is expected that apoptosis of cancer cells is induced with minimal damage to the normal epithelium.
It is reported that all prostate cancer cells express ganglioside GM2 on the cell surface, but GM1a synthesized from GM2a is negligible.43 Ganglioside GD1a is synthesized from GM1a by α 2,3 sialyltransferase (ST3Gal) I and/or II.45 RT-PCR analysis showed that the expression of ST3Gal I was much higher in PC3 cells than in LNCap and PNT2 (data not shown). The promoter analysis of STGal I gene revealed that the region from 146 to 304 is involved in the activation of ST3Gal I.46 STGal II gene has 2 alternative promoters (p1 and p2) and the regions from 661 to 801 of p1 promoter and from 1 to 161 of p2 promoter are activated in PC3 cells.47 SPG, a neo-lactoseries ganglioside, has not been analyzed in prostate cancers. Here, we first identified the increase of SPG in hormone-resistant prostate cancer cells (Fig. 2b). This is also a receptor for HVJ.32 The terminal sialic acid of SPG is transferred by ST3Gal VI.45 Human ST3Gal VI gene is expressed by alternative promoter utilization and one of the 2 promoters is specifically activated in PC3 cells.48 The promoter contains putative binding sites for various transcription factors.48 In PC3 and DU145 cells, both SPG and GD1a were highly produced (Fig. 2b). This result suggests that the activation of transcription factors required for both STGal I (or II) and STGal VI may result in the production of both GD1a and SPG. Potential transcription factors are USF, GATA-1, AML-1a, AP-1 and Nkx-2. Among these factors, AP-1 may be the most relevant to the progression of prostate cancers as recently reported.49, 50 Therefore, our hypothesis is that up-regulated AP-1 activates the expression of both STGal I/II and VI in hormone-independent prostate cancers, which results in high production of GD1a and SPG, receptors for HVJ. In the treatment of prostate cancers, hormone therapy is initially selected. When the cancers turn resistant to hormone therapy, cell surface gangliosides are also altered to be abundant in GD1a and SPG. Then, HVJ-E treatment should be chosen to efficiently eradicate hormone-resistant prostate cancers.
Live HVJ has been used to examine IFN secretion after infection. We have previously reported that HVJ-E also induces IFN secretion from DCs through a membrane-fusion dependent mechanism.40 The production of IFN by double-stranded RNA can be mediated by both RIG-I and TLR-3.39, 41 With HVJ-E, membrane fusion occurs at the cell surface, not in endosomes.22 The fact that gene delivery by HVJ-E is not affected by inhibitors of endocytosis51, 52 indicates that the viral genome can be directly delivered to the cytoplasm. Therefore, we theorize that IFN production by HVJ-E is also mediated by RIG-I in the cytoplasm, not by TLR3, which is present on the endosomal membrane. We also discovered up-regulation of RIG-I helicase by HVJ-E treatment. This is presumably because of the autocrine effect of secreted IFN as previously reported.41 In fact, RIG-I was activated upon the addition of recombinant IFN to PC3 cells (data not shown), and the effect of HVJ-E on RIG-I up-regulation was abolished with use of an anti-IFN receptor antibody (Fig. 4b). It is likely that up-regulation of RIG-I can augment the anti-tumor activity of HVJ-E treatment.
In Table I, several interferon-inducible antiviral proteins were up-regulated by HVJ-E treatment. They are 2′-5′-oligoadenylate synthetase (OAS), myxovirus resistance 1 (MX1), and hect domain and RLD5 (HERC5). In OAS family, only OAS 1 is activated by viral RNA to convert inactive RNaseL to active form.33 Active RNaseL nonspecifically digests cellular RNA as well as viral RNA to induce cell death. MX1 traps influenza virus proteins in the cytoplasm and induces degradation of the virus.33 HERC 5 is a ubiquitin ligase enzyme and forms ligation complex with E1-activating enzyme ISG15 and E2-conjugating enzyme UBCH8. The complex induces protein degradation. Therefore, at least OAS1 and HERC5 may promote cancer cell death.33 Moreover, sterile α motif domain containing 9 (SAMD9) is known to induce apoptosis in cancer cells.34 Interferon-induced protein 44 (IFI44) induces cell cycle-arrest.35 Interferon induced transmembrane protein 1 (IFITM1) and interferon-induced protein with tetratricopeptide repeats 3 (IFIT3, also called RIG-G) are reported to generate anti-proliferation activity.36, 37 These 3 proteins presumably assist cancer cell death. Interferon-induced with helicase C domain 1 (IFIH1) and DEAD box polypeptide 58 (DDX58) are also called MDA-5 and RIG-I that are cytoplasmic receptors of cytomegalovirus and Sendai virus, respectively.39 As described in the Results section, RIG-I expression is elevated by HVJ-E treatment. It may amplify anti-tumor activity of HVJ-E. Interferon-stimulated transcription factor 3 (ISGF3) inhibits IL-8 expression, which may suppress angiogenesis of tumor vessels.38 Dehydrogenase/reductase member 2 (DHRS2) is not interferon-inducible. The function of the enzyme doesn't seem to be related with anti-tumor activity.53 Thus, many genes activated with HVJ-E are interferon-inducible and appears to promote anti-tumor activity of HVJ-E.
In the present experiment, HVJ-E activated caspase 8, not caspase 9, induces apoptosis in PC3 cells. As p53 is known to be deleted in PC3 cells,54 it is likely that release of cytochrome c from the mitochondrial membrane is limited, which may result in failure of caspase 9 activation.55 As shown in Figure 5a, a JAK inhibitor reduced the activation of caspase 3 and 8 to control levels. This indicates that activation of caspase 8 was mediated by the JAK/STAT pathway, which is the classical mechanism of caspase 8 activation.42 Because the anti-IFN receptor antibody blocked both the phosphorylation of STAT1 and caspase activation, activation of the JAK/STAT pathway by HVJ-E appears to be mediated by the secretion of IFN following HVJ-E treatment.
When HVJ-E was added to cultured PC3 and DU145 cells, numerous polynuclear cells were generated as a result of cell-to-cell fusion induced by HVJ-E. Fused cells were not observed in either LNCap cells or PNT2 cells upon exposure to HVJ-E. The observed specificity of cell fusion in PC3 and DU145 cells upon exposure to HVJ-E is likely because of the expression of GD1a, a receptor for HVJ, which predominates in hormone-resistant prostate cancer cells.43, 44 Because fused cells stop dividing and became apoptotic, selective fusion of cancer cells may contribute to the anti-tumor effects of HVJ-E, in addition to the tumor killing activity of IFN secretion as well as the induction of anti-tumor immunity.
We have previously reported that HVJ-E evokes anti-tumor immunity in tumor-bearing mice following repeated intra-tumor injection.26–28 It induces tumor-specific NK cell migration through CXCL10 production by DCs, and enhances NK cell activity through the secretion of IFN from DCs.28 HVJ-E also enhances DC maturation and activates CTLs against cancer cells by enhancing effector T cells and suppressing regulatory T cells.27 In the present study, the most of intra-dermal PC3 cell tumor masses were completely eradicated by HVJ-E injection. The remarkable anti-tumor activity of HVJ-E was likely because of Type I IFN secretion by tumor cells, the formation of fused cells, as well as enhanced NK cell activation as suggested in Figure 6c. Tumor regression by HVJ-E was slightly but significantly attenuated by the anti-asialo GM1 antibody on days 60 and 70 (p < 0.05). However, significant suppression of tumor growth was still observed by HVJ-E treatment even in the presence of anti-asialo GM1 antibody compared with the group without HVJ-E (Fig. 6c). It is also thought that CTLs are generated against prostate cancer cells following HVJ-E injection in immune-competent mice as seen in previous reports.27, 56
HVJ-E was originally developed as a drug delivery vector.25 Therapeutic molecules, including DNA, siRNA, proteins and anti-cancer drugs, can be efficiently delivered to various cell lines both in vitro and in vivo.26 When bleomycin is delivered to cancer cells using HVJ-E, most cancer cells are killed by 0.33 μg/ml bleomycin, whereas 50% of cancer cells remain when 100 μg/ml bleomycin is used in the absence of HVJ-E.56 More than 80% of the tumors derived from CT26 cells are eradicated after 3 injections of bleomycin-incorporated HVJ-E in mice, which is associated with the generation of a large number of CTLs against CT26 cells. Thus, HVJ-E can treat cancers by delivering therapeutic molecules with the enhancement of anti-tumor immunity. In the treatment of hormone-resistant prostate cancer, HVJ-E alone seems to effectively eradicate primary tumors. However, combining these modalities will further augment the effects of HVJ-E in clinical trials.
Patients with hormone-refractory prostate cancers have multiple metastasis. The treatment of those metastatic tumors has been a big problem in cancer therapy. For the treatment of multiple metastasis, much attention has been paid to tumor-targeting vectors. We have already reported the technology to develop tissue-targeting HVJ-E.57 We have also developed the technology to establish HN-depleted HVJ-E using HN-specific siRNA.58 By combining these 2 technologies, tumor-targeting HVJ-E has been prepared.59 Antibody against prostate-specific membrane antigen or prostate stem cell antigen60, 61 may be useful for constructing prostate cancer targeting HVJ-E based on our technologies. The prostate cancer-targeting HVJ-E will apply to treat patients with multiple metastatic prostate cancers after strict safety tests in non-human primates. From the standpoint of tissue-targeting, the application of current HVJ-E to cancer treatment appears to be restricted. However, current HVJ-E also has potentiality to treat metastatic prostate cancers. One advantage of HVJ-E is to evoke T cell immunity against cancers. HVJ-E efficiently activates CTLs against cancers by the induction of DC maturation, activation of both CD4+ and CD8+ lymphocytes recognizing tumor-antigens and the suppression of regulatory T cells.27 Therefore, it is likely that local injection of HVJ-E to residual cancer tissues and regional lymph nodes activates systemic immune response against cancers to eliminate other untreated metastatic tumors. To test the possibility, clinical trials for the treatment of hormone-refractory prostate cancers using a clinical grade HVJ-E26 will be started soon.