BHK21 cells (ATCC Cat. No. CCL-10) and explanted splenocytes were maintained in RPMI-1640 containing 2 mmol/L L-glutamine, 1mmol/L sodium pyruvate, nonessential amino acids, 100 IU penicillin/streptomycin, 50μmol/L β-mercaptoethanol and 10% fetal bovine serum in a 37°C incubator with 5% CO₂. The HPV-16 E7-expressing murine tumor model TC-1 cell line, and its luciferase-expressing derivative line TC-1/Luciferase have been previously described (Lin et al., 1996; Huang et al., 2007). Briefly, primary lung epithelial cells of C57BL/6 mice were transformed with HPV-16 E6, E7, and an activated ras oncogene. These cells were then transduced with viral particles packaged with a Luciferase/Thy1.1 expression construct on the pMSCVpuro backbone (a kind gift from Hyam Levitsky). TC-1/Luciferase cells were cultured in the same media as BHK21 cells. Phycoerytherin-conjugated anti-mouse FasL antibody (clone MFL3; eBioscience, San Diego, CA) and isotype control (clone eBio299Arm, eBioscience) were used for cell surface staining of FasL, which was performed according to the vendor’s protocol. Phycoerythrin-conjugate monoclonal rat anti-mouse CD8 antibody and FITC-cojugated anti-mouse IFN-γ antibodies (BD Pharmingen, San Diego, CA) were used for intracellular cytokine staining of explanted splenocytes. For the FasL blocking experiments, antibodies used were the anti-mouse FasL (CD178.1) Ab (Kay-10 clone) (BioLegend, San Diego, CA) and the functional grade mouse IgG2b isotype control (eBioscience, San Diego, CA). The DC-1 dendritic cell line was kindly provided by Dr. Kenneth Rock (University of Massachusetts, Worcester, MA) (Kim et al., 2004).

Construction of the plasmid encoding Firefly luciferase (pcDNA3-Luc), as well as that encoding HPV-16 Sig/E7/LAMP-1 has been described previously. The shRNA constructs pRS-Luc (Cat. No. TR30002) and pRS empty vector control (Cat. No. TR20003) were purchased from OriGene (Rockville, MD). The FasL-GFP expression vector and GFP empty vector control were kind gifts from Drs. Linzhao Cheng and Xiaobing Yu. FasL Mission^™ shRNA TRCN0000066640 (CCGGCTTCGTGTATTCCAAAGTATACTCGAGTATACTTTGGAATACACGAAG TTTTTG) and the Mission^™ non target shRNA control vector (Cat. No. SHC002) (CCGGCGTGATCTTCACCGACAAGATCTCGAGATCTTGTCGGTGAAGATCACG TTTTT) were purchased from Sigma-Aldrich (St. Louis, MO). DNA plasmids were amplified in Escherichia coli DH5α (pcDNA3-Luc, pRS vector, pRS-Luc, and Sig/E7/LAMP-1) or STBL2 (FasL-GFP, GFP-empty vector, FasL Mission^™ shRNA) and purified as reported by Chen et al. using an endotoxin-free plasmid purification kit (Qiagen, Valencia, CA) (Chen et al., 2000).

For the experiments with luciferase shRNA, BHK-21 cells were seeded in a 24-well plate at a density of 5×10⁴ cells per well the day before transfection. Lipofectamine 2000 (Invitrogen, Carlsbad, CA) was used to transfect the cells with DNA plasmids (0.01 μg of pcDNA3-Luc, combined with a 0.81 μg mixture of pRS-Luc and pRS vector) in quadruplicate for each Luc shRNA dose. Cells were washed, and culture media was replaced subsequent to overnight incubation and daily luciferase-based bioluminescence imaging. For the experiments with FasL shRNA, BHK-21 cells were seeded in 10 cm dishes at 1.5×10⁶ cells per dish. Eight hrs later, FasL-GFP expression vector (2.4 μg) combined with FasL shRNA or non-specific shRNA construct (21.6 μg each) were transfected into the cells using Lipofectamine 2000. Cell surface expression of FasL was assayed following overnight incubation.

BHK-21 cells transfected with combinations of FasL-GFP expression vector and FasL shRNA were scraped after 24 hrs incubation at 37 °C, resuspended in FACscan buffer, and surface-stained in the dark with phycoerythrin-conjugated anti-mouse FasL antibody or isotype control for 30 min at 4°C. FasL expression analysis was performed on a Becton-Dickinson FACScan flow cytometer with CELL Quest software (Becton Dickinson Immunocytometry System, Mountain View, CA).

DNA-coated gold particles were prepared according to a previously described protocol (Chen et al., 2000), and delivered to the shaved abdominal region of C57BL/6 mice using a helium-driven gene gun (BioRad, Hercules, CA) with a discharge pressure of 400 p.s.i. For the kinetic measurement of epidermal luciferase expression, two DNA bullets were administered to non-contiguous areas per mouse (1 μg pcDNA3-Luc or pcDNA3 vector per bullet). For measurement of luciferase shRNA activity in vivo, one DNA bullet was administered per mouse (0.5 μg pcDNA3-Luc combined with 0.5 μg of pRS-Luc or pRS vector; 1 μg total DNA/bullet). For the FasL shRNA experiments, mice were primed with two bullets of DNA (0.1μg of Sig/E7/LAMP-1 combined with 0.9 μg of FasL shRNA or negative control shRNA per bullet; 2 μg total DNA/mouse) and boosted with the same dose one week later. For the FasL shRNA tumor treatment experiments, mice were vaccinated with two bullets of DNA (0.1μg Sig/E7/LAMP-1 combined with 0.9μg FasL shRNA or negative control shRNA per bullet, as per immune response experiment) eight days after tumor inoculation, and boosted three times with the same dose in four-day intervals.

The expression of luciferase in cultured BHK-21 cells was detected using the Xenogen IVIS 200 system (Xenogen Corp., Alameda, CA) and quantitated with the Living Image software package (Xenogen Corp.) Briefly, cells were washed twice with PBS, then 1 ml of media with 39 μg/ml beetle luciferin (potassium salt; Promega Corp., Madison, WI) was added to the cells and followed immediately by bioluminescence imaging (1 sec exposure). The expression of luciferase in the skin of mice was visualized with the same apparatus and software. C57BL/6 mice were injected intraperitoneally with 0.2ml of 3.9mg/ml beetle luciferin in PBS and placed under general anesthesia using isoflurane for 10 min. Luciferase activity was recorded with the IVIS digital camera (30 sec exposure).

Splenocytes were harvested one week after the last vaccination from mice (3 per group) immunized with Sig/E7/LAMP-1 DNA in combination with DNA coding for FasL shRNA or nonspecific shRNA. 5×10⁶ pooled splenocytes from each vaccination group were incubated overnight at 37°C with 1 μg/ml E7 peptide (RAHYNIVTF) in the presence of 1 μg/ml GolgiPlug (BD Pharmingen, San Diego, CA). Cells were then washed with FACScan buffer and co-stained with phycoerythrin-conjugated monoclonal rat anti-mouse CD8 antibody as well as FITC-conjugated IFN-γ antibody using the Cytofix/Cytoperm kit according to the manufacturer’s instructions (BD Pharmingen). Analysis was performed on a Becton-Dickinson FACScan flow cytometer with CELL Quest software.

Splenocytes were harvested from mice immunized with Sig/E7/LAMP-1 DNA in combination with DNA coding for FasL shRNA or non-specific shRNA (3 per group) one week after the last vaccination. Splenocytes of mice from each treatment group were incubated in pools of five million cells with 1 μg/ml of E7 peptide (RAHYNIVTF) and 20 units/ml IL-2 for five days, then collected and combined with TC-1/Luciferase target cells in 10:1 effector:target ratios. After adjusting cell volumes by centrifugation and resuspension in equal volumes of media (to achieve 5×10⁴ target cells/100 μl), cell mixtures were plated in 100μl quadruplicate aliquots to 96-well round-bottom plates, lightly centrifuged to bring cell mixtures to the bottom of wells, and incubated overnight. Following incubation, media was drawn from wells, briefly centrifuged to remove cells and cell debris, then quantitatively assayed in quadruplicate for luciferase activity by addition of Steady-Glo reagent (Promega, Madison, WI) in light-resistant black plastic 96-well plates. Visualization and signal quantitation was performed using the Xenogen IVIS system, and Living Image software package.

Mice (five per group) were challenged with 5×10⁴ TC-1/Luciferase tumor cells/mouse by subcutaneous injection in the right flank. DNA vaccine treatments (2μg/dose) were initiated seven days after tumor inoculation, and boosted three times with the same dose in four-day intervals. Tumor growth was periodically visualized under general anesthesia using the Xenogen IVIS system, and quantitated using the Living Image software package. Mice were also observed for signs of morbidity (cachexia, failure to groom, decreased social interaction with cage-mates), and euthanized when overtly moribund.

2 × 10⁵ FasL-blocked E7-specific CD8⁺ T cells were incubated with or without 1 × 10⁷ E7 peptide-loaded DC-1 cells transfected with FasL-GFP at a 1:50 ratio in the presence of the 5μg/ml of FasL blocking Ab or isotype control. The percentage of apoptotic cells were characterized using intracellular staining with antibodies specific for activated caspase-3 and analyzed by flow cytometry analysis.

We performed in vitro bioluminescence imaging to characterize luciferase expression in BHK-21 cells transfected with DNA encoding luciferase in combination with various amounts of DNA encoding shRNA targeting luciferase. As shown in Figure 2A, expression of luciferase declined with increasing dose of DNA encoding shRNA targeting luciferase, indicating the silencing of luciferase gene increased in a dose-dependent manner. Luciferase activity becomes virtually undetectable in cells transfected with more than 0.09 μg of DNA encoding luciferase-specific shRNA. We further characterized the kinetics of silencing of the luciferase gene expression over a period of 5 days in cells transfected with different amounts of DNA encoding luciferase-specific shRNA. We observed a dose-dependent silencing of luciferase gene expression over time. Furthermore, luciferase activity remained substantially suppressed even after 120 hours (Figure 2B). Thus, these data demonstrate that luciferase can be continuously silenced using DNA encoding shRNA targeting luciferase in a dose-dependent manner in vitro.

We next sought to determine whether DNA encoding shRNA targeting luciferase could be intradermally administered by gene gun to knock down luciferase expression in vivo as assayed by noninvasive bioluminescence imaging. C57BL/6 mice (5 per group) were intradermally administered with DNA encoding luciferase and DNA encoding shRNA targeting luciferase (pcDNA3-Luc + pRS-Luc) or DNA encoding luciferase and empty vector DNA (pcDNA3-Luc + pRS-empty). The luciferase activity was monitored over time using bioluminescence imaging. As shown in Figure 3A, representative bioluminescence images of mice administered with DNA encoding luciferase, combined with DNA encoding luciferase-specific shRNA, displayed significantly lower levels of luciferase expression compared to mice administered with DNA encoding luciferase combined with empty vector DNA at 24 hours. Furthermore, Figure 3B indicates the silencing of luciferase expression over time. We observed that the luminescent intensity in mice administered with pcDNA3-Luc + pRS-Luc was significantly lower compared to mice administered with pcDNA3-Luc + pRS-empty vector at 24 hours after DNA administration (~3-fold difference, p < 0.01). These results show that DNA encoding shRNA targeting luciferase can be efficiently administered intradermally with the gene gun to achieve efficient silencing of luciferase in vivo. These encouraging data suggest that we may be able to use intradermal administration via gene gun to modify the properties of DCs using DNA encoding shRNA targeting immunosuppressive molecules to enhance DNA vaccine potency.

The apoptosis inducer FasL, a protein abundantly expressed on the surface of DCs, may serve as an ideal target for gene silencing to enhance DNA vaccine potency by potentially reducing DC-induced T cell death (Lu et al., 1997). During antigen presentation, FasL expressed by DCs may induce apoptosis in some T cells through association with the Fas death receptor present on the surface of T cells, concomitant with MHC-T cell receptor interactions at the immunological synapse. In order to determine whether transfection of FasL-expressing cells with DNA encoding shRNA targeting FasL could lead to downregulation of FasL in vitro, we transfected BHK-21 cells with plasmid DNA encoding FasL and DNA encoding shRNA specific for FasL. We then performed flow cytometry analysis to characterize the expression of FasL in transfected cells. As shown in Figure 4, cotransfection of cells with DNA encoding FasL and DNA encoding FasL-specific shRNA led to significant reduction of FasL expression compared to the cells cotransfected with DNA encoding non-specific shRNA (p < 0.05 by Student’s t test). Thus, our data indicate that transfection of FasL-expressing cells with DNA encoding shRNA specific for FasL substantially downregulates FasL expression in vitro.

We have previously generated a DNA construct containing the HPV-16 E7 linked to the sorting signal of the lysosome-associated membrane protein (LAMP-1) to form a chimeric protein Sig/E7/LAMP-1. We found that intradermal administration of DNA encoding Sig/E7/LAMP-1 could generate enhanced E7-specific CD8⁺ T cell immune responses in vaccinated mice (Chen et al., 1999). We then sought to determine whether the intradermal administration of Sig/E7/LAMP-1 DNA could be further enhanced by co-administration with DNA encoding shRNA targeting FasL. C57BL/6 mice (3 per group) were vaccinated intradermally via gene gun with Sig/E7/LAMP-1 DNA and DNA encoding shRNA targeting FasL, and boosted with the same dose one week later. One week after the last vaccination, splenocytes were harvested and characterized for E7-specific CD8⁺ T cells by intracellular cytokine staining followed by flow cytometry analysis. As shown in Figure 5A, a significantly enhanced E7-specific CD8⁺ T cell immune response was observed in mice vaccinated with Sig/E7/LAMP-1 DNA and DNA encoding FasL shRNA compared to mice vaccinated with Sig/E7/LAMP-1 DNA and DNA encoding nonspecific shRNA. A graphical representation of the number of E7-specific IFN-γ⁺ CD8⁺ T cells in each group is depicted in Figure 5B (p < 0.005 by Student’s t test). Thus, our data indicate that intradermal administration of Sig/E7/LAMP-1 DNA with DNA encoding shRNA targeting FasL results in a significant increase in the number of E7-specific CD8⁺ T cells in vaccinated mice and suggests that RNAi-mediated knockdown of FasL expression in DCs may represent an effective strategy for augmenting vaccine potency.

To evaluate if the increase in the number of E7-specific CD8⁺ T cells generated by vaccination of Sig/E7/LAMP-1 DNA vaccines in combination with FasL shRNA translate into a potent cytotoxic activity against E7-expressing tumors, we performed in vitro cytotoxicity assays using in vitro stimulated splenocytes from vaccinated C57BL/6 mice with a luciferase expressing TC-1 tumors (TC-1/luciferase). An E7-specific T cell line was used as a positive control and TC-1/Luciferase tumor cells without T cells were used as a negative controls. As shown in Figure 5C, splenocytes obtained from animals vaccinated with Sig/E7/LAMP-1 DNA vaccines in combination with DNA encoding FasL shRNA demonstrated significantly enhanced cytolytic activity compared to splenocytes obtained from animals vaccinated with Sig/E7/LAMP-1 DNA vaccines with DNA encoding non-specific shRNA (p<0.001). Thus, our data suggest that the enhanced E7-specific CD8⁺ T cells observed in mice vaccinated with Sig/E7/LAMP-1 DNA with DNA encoding FasL-specific shRNA also demonstrated leads potent cytolytic activity against an E7-expressing model tumor cell line.

In order to determine the therapeutic anti-tumor effects of intradermal administration of Sig/E7/LAMP-1 DNA with DNA encoding FasL shRNA, we performed in vivo tumor treatment experiments. C57BL/6 mice (5 per group) were challenged with 5×10⁴ TC-1/Luciferase tumor cells/mouse by subcutaneous injection in the right flank, and monitored for tumor growth over time by non-invasive bioluminescence imaging. The luminescent intensity has been shown to correlate with tumor load (Hung et al., 2007a). Eight days after tumor challenge, tumor-bearing mice were treated with Sig/E7/LAMP-1 DNA vaccine combined with DNA encoding FasL shRNA, or non-specific shRNA via gene gun. The mice received three boosters with the same dose intradermally at four-day intervals. As shown in Figure 6, while untreated tumor-bearing mice demonstrated continued increase in the tumor luminescence intensity over time (p=0.036), mice treated with Sig/E7/LAMP-1 DNA vaccine and DNA encoding FasL shRNA demonstrated continued decrease in the tumor luminescence intensity over time (p=0.042). Furthermore, mice treated with Sig/E7/LAMP-1 DNA vaccine and DNA encoding non specific shRNA demonstrated some therapeutic effect as shown by the decrease in tumor luminescence intensity over time, but the decrease in tumor luminescence intensity was not statistically significant (p=0.184). Taken together, our results suggest that the treatment with the DNA vaccine combined with DNA encoding FasL-specific shRNA generates potent therapeutic anti-tumor effects against E7-expressing tumors.