Enhancement of DNA Vaccine Potency by Linkage of Antigen Gene to a Gene Encoding the Extracellular Domain of Fms-like Tyrosine Kinase 3-Ligand¹

Antigen-specific cancer immunotherapy has emerged as a promising approach for controlling cancer because it is capable of developing specific immunity against neoplastic cells while not attacking normal cells. Increasing evidence suggests that professional APCs,⁴particularly DCs, are the central players for mediating cancer immunotherapy. An effective vaccine most likely requires a strategy that targets antigen to professional APCs to activate antigen-specific T cells (reviewed in Ref. 1).

Recently, DNA vaccines have become an attractive approach for generating antigen-specific immunotherapy (reviewed in Refs.2, 3, 4). One of the concerns about DNA vaccines is their potency, because they do not have the intrinsic ability to amplify in vivo as viral vaccines do. We reasoned that a DNA vaccine encoding a fusion antigen that is directed to cells which activate immune responses, such as DCs, may enhance vaccine potency. Previously,it has been demonstrated that the linkage of a GM-CSF gene to an antigen gene enhanced the potency of DNA vaccines against HIV(5) and hepatitis C (6). It is believed that chimeric GM-CSF/antigen may act as an immunostimulatory signal to DCs,inducing differentiation from an immature DC form to a mature form(7). Because DCs and their precursor cells express high levels of GM-CSF receptors, chimeric GM-CSF/antigen may target and concentrate the linked antigen to DCs and further improve DNA vaccine potency.

Another important molecule that also possesses a growth-stimulatory effect on DC precursors and has been shown to be capable of generating large numbers of DCs in vivo is FL (8, 9). FL has emerged as an important molecule for the development of tumor vaccines that augment the function and quantity of DCs in vivo. Flt3, a murine tyrosine kinase receptor, was first described in 1991 (10) and was found to be a member of the same family of receptors as c-kit and c-fms receptors, the type III receptor kinase family (reviewed in Ref. 11). In hematopoietic tissues, the expression of Flt3 is restricted to the CD34-positive progenitors. Flt3 has been used to identify and subsequently clone the corresponding ligand, FL (12, 13).

The predominant form of FL is synthesized as a transmembrane protein from which the soluble form is generated, presumably by proteolytic cleavage. The soluble form of FL (extracellular domain) has been shown to be functionally similar to FL (12). These proteins function by binding to and activating unique tyrosine kinase receptors. Expression of the Flt3 receptor is primarily restricted, among hematopoietic cells, to the most primitive progenitor cells, including DC precursors. Several studies have shown that the soluble extracellular domain of FL generated strong antitumor effects against several murine model tumors including fibrosarcoma (14),breast cancer (15, 16), liver cancer (17),lung cancer (18), melanoma, and lymphoma(19). To date, FL has not been used in the form of chimeric DNA vaccines.

In our current study, we investigated whether linking a full-length E7 gene to a gene encoding the extracellular domain of FL would enhance the potency of DNA vaccines. We chose human HPV-16 E7 as a model antigen for vaccine development because HPVs, particularly HPV-16, are associated with most cervical cancers. The HPV oncogenic proteins, E6 and E7, are important in the induction and maintenance of cellular transformation and are coexpressed in most HPV-containing cervical cancers. Vaccines or immunotherapies targeting E7 and/or E6 proteins may provide an opportunity to prevent and treat HPV-associated cervical malignancies. We compared DNA vaccines containing wild-type HPV-16 E7 with DNA vaccines containing full-length E7 fused to FL for their generation of immune responses and their ability to protect animals against HPV-16 E7-expressing murine tumors (20). Our data indicated that linkage of a gene encoding the extracellular domain of FL to E7 dramatically increases the expansion and activation of E7-specific CD8⁺ T cells, completely bypassing the CD4 arm. This strategy enhanced E7-specific CD8⁺ T-cell responses, resulting in potent antitumor immunity against established E7-expressing metastatic tumors.

We used pcDNA3 as an expression vector instead of a previously described pCMV-Neo-Bam vector (21, 22). We observed that the pCMV-Neo-Bam vector generated a strong antitumor effect but weak E7-specific T cell-mediated immune response in mice vaccinated with wild-type E7 DNA (21, 22). The degree of antitumor effect generated by DNA vaccines using pCMV-Neo-Bam vector did not appear to correlate well with E7-specific T cell-mediated immune activity. We chose to use the pcDNA3 expression vector, which has been used effectively to investigate the correlation between the E7-specifc T cell-mediated immune response with the antitumor effect generated by various DNA vaccines (23). The generation of HPV-16 E7-expressing plasmid, pcDNA3-E7, has been described previously(23). For the generation of plasmid encoding the extracellular domain of mouse FL, pcDNA3-FL, the DNA fragment encoding the signal peptide and extracellular domain of mouse FL was first amplified with PCR using conditions as described previously(23) with a mouse FL DNA template, sfHAV-EO410 (American Type Culture Collection, Manassas, VA) and a set of primers:5′-gggtctagaatgacagtgctggcgccagc-3′ and 5′-gggggatccctgcctgggccgaggctctgg-3′. The amplified product was then digested with XbaI and BamHI and further cloned into the XbaI and BamHI cloning sites of pcDNA3 vector (Invitrogen, Carlsbad, CA). For the generation of pcDNA3-FL-E7,the E7 DNA fragment was isolated from pcDNA3-E7 by digestion with BamHI and HindIII and gel recovered. The isolated E7 DNA fragment was further cloned into the BamHI and HindIII cloning sites of pcDNA3-FL. For the generation of pcDNA3-GFP, a DNA fragment encoding the GFP was first amplified with PCR using pEGFPN1 DNA (Clontech, Palo Alto, CA) and a set of primers:5′-atcggatccatggtgagcaagggcgaggag-3′and 5′-gggaagctttacttgtacagctcgtccatg-3′. The amplified product was then digested with BamHI and HindIII and further cloned into the BamHI and HindIII cloning sites of pcDNA3 vector (Invitrogen). For the generation of pDNA3-E7-GFP, a DNA fragment encoding HPV-16 E7 first was amplified with PCR using pcDNA3-E7 as template and a set of primers:5′-ggggaattcatgcatggagatacaccta-3′ and 5′-ggtggatccttgagaacagatgg-3′. The amplified product was then digested with EcoRI and BamHI and further cloned into the EcoRI and BamHI cloning sites of pcDNA3-GFP vector. For the generation of pcDNA3-FL-E7-GFP, the DNA encoding the signal peptide and extracellular domain of FL was amplified with PCR using pcDNA3-FL as a DNA template and a set of primers: 5′-gggtctagaatgacagtgctggcgccagc-3′and 5′-cgagaattcctgcctgggccgaggctctg-3′. The amplified product was then digested with XbaI and EcoRI and further cloned into the XbaI and EcoRI cloning sites of pcDNA3-E7-GFP vector. The accuracy of these constructs was confirmed by DNA sequencing. pcDNA3 DNA with FL, E7, FL-E7, E7-GFP, or FL-E7-GFP gene insert and the “empty” plasmid, pcDNA3 vector, were transfected into subcloning-efficient DH5α cells (Life Technologies,Inc., Rockville, MD). The DNA was then amplified and purified as described previously (23). The integrity of plasmid DNA and the absence of Escherichia coli DNA or RNA was checked in each preparation using 1% agarose gel electrophoresis. DNA concentration was determined by the absorbance measured at 260 nm. The presence of inserted E7 fragment was confirmed by restriction enzyme digestion and gel electrophoresis.

The production and maintenance of TC-1 cells has been described previously (20). On the day of tumor challenge, TC-1 cells were harvested by trypsinization, washed twice with 1× HBSS, and finally resuspended in 1× HBSS to the designated concentration for injection. A human embryonic kidney 293 cell line expressing the D^b and K^b (293 D^b,K^b; Ref.24) was a gift from Dr. J. C. Yang (National Cancer Institute, NIH, Bethesda, MD). It was grown in DMEM medium containing 10% heat-inactivated FCS, 0.3% glutamine, 0.01 m HEPES,100 units/ml penicillin, 100 μg/ml streptomycin, and 400 μg/ml G418.

293 D^b,K^b cells transfected with pcDNA E7-GFP and pcDNA FL-E7-GFP DNA were cultured for 24–36 h and then cytospinned to glass slides. Cells were fixed with 4%paraformaldehyde in 1× PBS for 30 min at room temperature,permeabilized with 1× PBS containing 0.05% saponin and 1% BSA, and then incubated with mouse anti-calnexin MAb (Stressgen Biotechnologies,Victoria, British Columbia, Canada) at a concentration of 1μg/ml for 30 min at room temperature. Unbound antibodies were removed by washing three times in 1× PBS. The cells were then incubated with Cy3-conjugated F(ab′)2 fragment goat antimouse IgG (Jackson ImmunoReseach Laboratories) at the concentration of 10 μg/ml for 30 min. The slides were washed with 1× PBS containing and 1% BSA. The glass slides were mounted with anti-fading medium, Mowiol 4–88(Calbiochem Inc., La Jolla, CA) and covered with coverslips. Slides skipping primary antibody were used as negative controls. Samples were examined on a confocal laser scanning microscopy.

Female C57BL/6 mice, 6–8 weeks of age from the National Cancer Institute (Frederick, MD), were purchased and kept in the oncology animal facility of the Johns Hopkins Hospital (Baltimore, MD). All animal procedures were performed according to approved protocols and in accordance with recommendations for the proper use and care of laboratory animals.

Preparation of DNA-coated gold particles and gene gun particle-mediated DNA vaccination was performed using a helium-driven gene gun (Bio-Rad,Hercules, CA) according to a protocol described previously(23). Briefly, DNA-coated gold particles (1 μg DNA/bullet) were delivered to the shaved abdominal region of the mice using a helium-driven gene gun (Bio-Rad, Hercules, CA) with a discharge pressure of 400 p.s.i.

Splenocytes from naïve or vaccinated groups of mice were incubated either with the E7 peptide (amino acids 49–57) containing the MHC class I epitope (25) for detecting E7-specific CD8⁺ T cell precursors or the E7 peptide (amino acids 30–67) containing the MHC class II peptide (26) for detecting E7-specific CD4⁺ T helper cell precursors. The E7 peptide was added at a concentration of 2 μg/ml for 20 h. Golgistop (PharMingen, San Diego, CA) was added 6 h before harvesting the cells from the culture. Cells were then washed once in FACScan buffer and stained with phycoerythrin-conjugated monoclonal rat antimouse CD8 or CD4 antibody (PharMingen). Cells were subjected to intracellular cytokine staining using the Cytofix/Cytoperm kit according to the manufacturer’s instructions (PharMingen). FITC-conjugated anti-IFN-γ antibody and the immunoglobulin isotype control antibody (rat IgG1) were all purchased from PharMingen. Analysis was done on a Becton Dickinson FACScan with CELLQuest software(Becton Dickinson Immunocytometry System, Mountain View, CA).

The anti-HPV 16 E7 antibodies in the sera were determined by a direct ELISA as described previously (27). Briefly, a 96-microwell plate was coated with 100 μl of 5 μg/ml bacteria-derived HPV-16 E7 proteins and incubated at 4°C overnight. The wells were then blocked with PBS containing 20% fetal bovine serum. Sera were prepared from the mice on day 14 after immunization,serially diluted in 1× PBS, added to the ELISA wells, and incubated at 37°C for 2 h. After washing with 1× PBS containing 0.05% Tween 20, the plate was incubated with 1:2000 dilution of a peroxidase-conjugated rabbit antimouse IgG antibody (Zymed, San Francisco, CA) at room temperature for 1 h. The plate was washed six times, developed with tetramethylbenzidine (Pierce Corp.,Rockford, IL), and stopped with 1 mH₂SO₄. The ELISA plate was read with a standard ELISA reader at 450 nm.

For the tumor protection experiment, mice (five/group) were vaccinated via gene gun with 2 μg of FL DNA, E7 DNA, FL-E7 DNA, or FL mixed with E7 (FL+E7), or they were unvaccinated. One week later, the mice were boosted with the same regimen as the first vaccination. One week after the last vaccination, mice were s.c. challenged with 1 × l0⁴ cells/mouse TC-1 tumor cells in the right leg and then monitored twice a week. Analysis was performed using SAS version 6.12 (SAS Institute Inc., Cary, NC). The percentage of tumor-free mice was analyzed according to Kaplan-Meier methods. Statistical significance was tested using log-rank statistics.

The tumor cells and DNA vaccines were prepared as described above. Mice were i.v. challenged with 1 × 10⁴cells/mouse TC-1 tumor cells via tail vein on day 0. Three days after challenge with TC-1 tumor cells, mice were given 2 μg of FL DNA, E7 DNA, or FL-E7 DNA via gene gun or unvaccinated. One week later, these mice were boosted with the same regimen as the first vaccination. The mice were sacrificed on day 25. The number of pulmonary metastatic nodules of each mouse was evaluated and counted by experimenters blinded to sample identity. Statistical significance was tested using one-way ANOVA.

In vivo antibody depletions have been described previously(20). Briefly, mice were vaccinated with 2 μg FL-E7 DNA via gene gun, boosted 1 week later, and challenged with 5 × 10⁴ cells/mouse TC-1 tumor cells. Depletions were started 1 week prior to tumor challenge. MAb GK1.5(28) was used for CD4 depletion, MAb 2.43(29) was used for CD8 depletion, and MAb PK136(30) was used for NK1.1 depletion. Flow cytometry analysis revealed that >95% of the appropriate lymphocyte subsets were depleted with normal levels of other subsets. Depletion was terminated on day 40 after tumor challenge.

DCs were generated by culture of bone marrow cells in the presence of GM-CSF as described previously (31). Briefly, bone marrow was collected from the femurs and tibias of mice. Erythrocytes were lysed, and the remaining cells were passed through a nylon mesh to remove small pieces of bone and debris. The cells were collected, and 1 × 10⁶ cells/ml were placed in 24-well plates in RMPI 1640, supplemented with 5% FCS, 2 mm β-mercaptoethanol, 1% nonessential amino acids,100 units/ml penicillin, and 100 μg/ml streptomycin (Life Technologies, Inc.), and 100 units/ml GM-CSF (PharMingen). Two-thirds of the medium was replaced every 2 days, and nonadherent cells were harvested on day 7. The collected cells were characterized by flow cytometry analysis for DC markers as described previously(32).

Generation of E7-specific CD8⁺ cell lines has been described previously (32). Briefly, female C57BL/6(H-2b) mice were immunized by i.p. injection of vaccina-Sig/E7/LAMP-1. Splenocytes were harvested on day 8. For initial in vitrostimulation, splenocytes were pulsed with IL-2 at a concentration of 20 units/ml and 1 μm E7 peptide (amino acids 49–57) for 6 days. Propagation of the E7-specific CTL cell line was performed in 24-well plates by mixing (2 ml/well)1 × 10⁶ splenocytes containing E7-specific CTLs with 3 × 10⁶ irradiated splenocytes and pulsing them with IL-2 at a concentration of 20 units/ml and 1 μm E7 peptide (amino acids 49–57). This procedure was repeated every 6 days. The specificity of the E7 CTL line was characterized by the CTL assay. Flow cytometry was performed to demonstrate the expression of the CD8 marker.

CTL assays were performed in 96-well round-bottomed plates as described by Corr et al. (33). Cytolysis was determined by quantitative measurements of LDH as reported previously(33). Transfected 293 D^b,K^b cells were used as target cells, whereas E7-specific CD8⁺ T cells served as effector cells. 293 D^b,K^b cells (5 × 10⁶) were transfected with 20 μg of pcDNA3 (empty plasmid), E7, FL, or FL-E7 DNA vaccines via Lipofectamine 2000 (Life Technologies, Inc.), according to the manufacturer’s protocol. The 293 D^b,K^bcells were collected 40–44 h after transfection. The levels of E7 protein expression as determined by ELISA were similar in E7- and FL-E7-transfected 293 D^b,K^b. CTL assays were performed with effector cells and target cells (1 × 10⁴ cells/well) mixed together at various ratios(1:1, 3:1, 9:1, and 27:1) in a final volume of 200 μl. After 5 h incubation at 37°C, 50 μl of the cultured medium were collected to assess the amount of LDH in the cultured medium using CytoTox assay kits (Promega Corp., Madison, WI) according to the manufacturer’s protocol. The percentage of lysis was calculated from the following equation: 100 × (A − B)/(C − D),where A is the reading of experimental-effector signal value, B is the effector spontaneous background signal value, C is maximum signal value from target cells, and D is the target spontaneous background signal value.

CTL assays using DCs pulsed with cell lysates as target cells were performed using a protocol similar to the protocol described by Uger and Barber (34). Briefly, 5 × 10⁶ 293 D^b,K^b cells were first transfected with 20 μg of pcDNA3 (empty plasmid), E7, FL, or FL-E7 DNA vaccines via Lipofectamine 2000 (Life Technologies, Inc.) according to the manufacturer’s protocol. The transfected 293 D^b,K^b cells were collected 40–44 h after transfection and then treated with three cycles of freeze-thaw. The protein concentration was determined using the Bio-Rad protein assay (Bio-Rad, Hercules, CA) using the vendor’s protocol. The quantity of E7 protein was determined using ELISA, and the cell lysates from E7- or FL-E7 DNA-transfected 293 D^b,K^b cells were standardized for E7 protein concentration. The DCs were used as target cells and prepared by pulsing 1 million of DCs with different concentrations of cell lysates (50, 10, 2, and 0.4 μg/ml) in a final volume of 2 ml for 16–20 h. E7-specific CD8⁺ T cells were used as effector cells. CTL assays were performed at fixed E:T (9:1) ratio with 9 × 10⁴ of E7-specific T cells mixed with 1 × 10⁴ of prepared DCs in a final volume of 200μl. Results from CTL assays were determined by quantitative measurements of LDH as described above.

To determine the expression and localization of wild-type and modified HPV-16 E7 protein, we have added the gene encoding GFP to the 3′ end of the E7 gene and chimeric FL-E7 genes as a tag. Transfection and subsequent examination with a fluorescence microscope were used to determine the expression and localization of wild-type and modified HPV-16 E7 protein. As shown in Fig. 1, the level of protein expression was quite similar between cells transfected with E7-GFP or FL-E7-GFP. As expected, cells transfected with E7-GFP showed cytoplasmic/nuclear distribution (Fig. 1,B). In comparison, cells transfected with the chimeric FL-E7-GFP construct displayed a network pattern consistent with ER localization (Fig. 1,E). To further explore whether the FL-E7-GFP chimera had in fact been distributed to the ER, we performed immunofluorescent staining of cells transfected with either E7-GFP or FL-E7-GFP using an antibody against calnexin (Fig. 1, A and D), a well-characterized marker for the ER. As shown in Fig. 1, C and F, colocalization of E7-GFP and the calnexin protein was observed only in cells transfected with FL-E7-GFP but not E7-GFP, indicating that at least some of the FL-E7 chimera was targeted to the compartments of the ER. These data indicated that the addition of the extracellular domain of FL to E7 may facilitate the entry of E7 into ER compartments.

CD8⁺ T lymphocytes are one of the most crucial effectors for inducing antitumor immunity. To determine the quantity of E7-specific CD8⁺ T-cell precursors generated by the FL-E7 DNA vaccine, we used intracellular cytokine staining, a sensitive functional assay for measuring IFN-γ production at the single-cell level (35). As shown in Fig. 2, mice vaccinated with FL-E7 DNA generated the highest number of E7-specific IFN-γ⁺ CD8⁺T-cell precursors (94/3 × 10⁵splenocytes) using flow cytometry analysis, whereas mice vaccinated with E7 DNA generated fewer precursors (12/3 × 10⁵ splenocytes). FL-E7 DNA immunization led to a nearly 8-fold increase in the number of E7-specific CD8⁺ T-cell precursors. These results also indicated that fusion of E7 to FL was required for enhancement of CD8⁺ T-cell activity, because FL mixed to E7(FL+E7 DNA) did not generate enhancement of CD8⁺T-cell activity.

To examine the generation of E7-specific CD4⁺ T precursor cells and cytokine profiles (IFN-γ or IL-4 secreting) by each of these vaccines, we performed double staining for CD4 surface marker and intracellular IFN-γ or IL-4 on splenocytes obtained from immunized mice, followed by flow cytometry analysis. As shown in Fig. 3,A, mice vaccinated with FL-E7 DNA generated no significant increase in CD4⁺ IFN-γ⁺double-positive cells compared with mice vaccinated with FL DNA,wild-type E7 DNA, plasmid DNA, or naïve mice without vaccination. Splenocytes from Sig/E7/LAMP-1 DNA vaccinated mice(22) were used as positive controls. Similarly, no significant increase in CD4⁺IL-4⁺ double-positive cells were identified in the mice vaccinated with FL-E7 compared with mice vaccinated with FL DNA, wild-type E7 DNA, plasmid DNA, or the naïve mice without vaccination (Fig. 3 B). IL-4-secreting activated mouse splenocytes (MiCK-2; PharMingen) were used as positive controls to assure the success of intracellular IL-4 staining for this study.

To determine the levels of E7-specific antibodies in the sera of the vaccinated mice, ELISA was performed 2 weeks after the last vaccination and read at 450 nm. No significant E7-specific antibody responses were detected in mice vaccinated with chimeric FL-E7 DNA, FL, or empty plasmid, or in naïve mice (data not shown).

To determine whether vaccination with various DNA vaccine constructs protects mice against E7-expressing tumors, in vivo tumor protection experiments were performed. As shown in Fig. 4, 100% of mice receiving FL-E7 DNA vaccination remained tumor-free 70 days after TC-1 challenge (log-rank, P < 0.001). In contrast, only 20% of mice receiving wild-type E7 remained tumor free after day 32, and all unvaccinated mice, or FL DNA-vaccinated mice, developed tumor growth within 20 days after tumor challenge. These results also indicated that fusion of E7 to FL was required for antitumor immunity, because only 20% of mice receiving FL mixed with E7 (FL+E7 DNA) remained tumor free after 32 days. Therefore,FL-E7 fusion DNA significantly enhanced the antitumor immunity against the growth of TC-1 tumors.

To determine the therapeutic potential of a chimeric FL-E7 DNA construct in treating TC-1 tumor metastases in the lungs, in vivo tumor treatment experiments were performed. As shown in Fig. 5,A, mice vaccinated with FL-E7 DNA demonstrated the lowest mean number of pulmonary nodules (5.8 ± 3.6) compared with mice vaccinated with wild-type E7 DNA (67.5 ± 3.5), FL mixed with E7 DNA (68 ± 15), or FL DNA(65 ± 5.0), or unvaccinated mice (50.66 ± 7.3; one-way ANOVA, P < 0.001). Data are expressed as mean number of pulmonary metastatic tumor nodules ± SE. Furthermore, mice vaccinated with FL-E7 DNA also had the lowest mean lung weight (g;0.158 ± 0.025) compared with mice vaccinated with wild-type E7 DNA (0.462 ± 0.02), FL plus E7 DNA(0.469 ± 0.08), or FL DNA (0.6 ± 0.03),or unvaccinated mice (0.645 ± 0.08; one-way ANOVA, P < 0.001; Fig. 5,B). Data are expressed as mean lung weight ± SE. Representative gross photographs of the lung tumors are shown in Fig. 6.

To determine the subset of lymphocytes that are important for the rejection of E7-positive tumor cells, we performed in vivoantibody depletion experiments (20, 36). Depletion of lymphocyte subsets was assessed on the day of tumor injection and weekly thereafter by flow cytometry analysis of spleen cells. More than 99% depletion of the appropriate subset was achieved with normal levels of the other subsets (data not shown). As shown in Fig. 7, all mice without vaccination and all of the mice depleted of CD8⁺ T cells grew tumors within 14 days after tumor challenge. In contrast, all of the nondepleted mice and all of the mice depleted of CD4⁺ T cells remained tumor free 60 days after tumor challenge. Forty % of NK1.1-depleted mice grew tumors 6 weeks after tumor injections. Although NK cells appear to be important for the antitumor effect compared with mice with no depletion, this difference was not statistically significant (log-rank, P = 0.13). Our results suggest that CD8⁺ T cells are essential for E7-specific antitumor immunity generated by the FL-E7 DNA vaccine.

From the immunological assays of vaccinated mice, we observed that mice vaccinated with FL-E7 generated the highest number of E7-specific CD8⁺ T-cell precursors (Fig. 2). To determine the mechanism that accounts for such a phenomenon, we first tested whether there was enhanced MHC class I presentation of E7 in cells expressing FL-E7 (in this case, human embryonic kidney 293 D^b,K^b cells transfected with FL-E7). We used CTL assays with D^b-restricted E7-specific CD8⁺ T cells as effector cells to determine whether target cells (293 D^b,K^b cells) transfected with FL-E7 can be killed more efficiently than 293 D^b,K^b cells transfected with wild-type E7. We chose 293 D^b,K^b cells as target cells because 293 D^b,K^b cells have been shown to have a stable high transfection efficiency(24). In addition, the level of E7 expression in FL-E7 or E7 DNA transfected 293 D^b,K^b cells was similar(data not shown). CTL assays were performed using 293 D^b,K^b cells transfected with empty plasmid, FL, E7, or FL-E7 DNA or nontransfected 293 D^b,K^b cells with various E:T ratios (1:1, 3:1, 9:1, and 27:1). As shown in Fig. 8, 293 D^b,K^b cells transfected with FL-E7 DNA generated significantly higher percentages of specific lysis compared with 293 D^b,K^b cells transfected with wild-type E7 DNA. These results indicated that cells transfected with FL-E7 DNA presented E7 antigen through the MHC class I pathway more efficiently than cells transfected with wild-type E7 DNA.

Another potential mechanism for the enhanced E7-specific CD8⁺ T-cell immune responses in vivois the presentation of E7 through the MHC class I pathway by uptake of lysed cells expressing various E7 DNA constructs, also called“cross-priming.” A cross priming experiment was performed to characterize the MHC class I presentation of E7 of DCs pulsed with cell lysates of 293 D^b,K^b cells transfected with empty plasmid, FL, E7, or FL-E7 DNA. Lysates of transfected 293 D^b,K^b cells were obtained from cycles of freeze-thaw. Bone marrow-derived DCs(1 × 10⁶) were pulsed with a serial dilution of lysates derived from various transfected 293 D^b,K^b cells (50, 10, 2, or 0.4 μg). DCs were used as target cells, whereas D^b-restricted E7-specific CD8⁺ T cells served as effector cells. CTL assays were performed with a fixed E:T ratio (9:1). As shown in Fig. 9, DCs pulsed with lysates from 293 D^b,K^b cells transfected with FL-E7 DNA generated significantly higher percentages of specific lysis as compared with DCs pulsed with lysates from 293 D^b,K^b cells transfected with the other DNA constructs and nontransfected DCs. These results suggested that DCs pulsed with FL-E7 fusion protein presented E7 antigen through the MHC class I pathway more efficiently than DCs pulsed with wild-type E7 protein. Thus, the fusion of FL to E7 may enhance E7-specific CD8⁺ T-cell immune responses via cross-priming effects.

In this study, we demonstrated that linkage of the extracellular domain of FL to E7 can significantly enhance the potency of HPV-16 E7-expressing DNA vaccines. DNA vaccines encoding FL fused to HPV-16 E7 generated significant CD8⁺ T cell-dependent preventive effects against HPV-16 E7-expressing murine tumors. Furthermore, the chimeric FL-E7 DNA vaccine was capable of controlling lethal pulmonary metastatic tumors.

Our data demonstrated that the incorporation of FL can preferentially enhance CD8⁺ T-cell responses of E7 DNA vaccines in vaccinated mice. In contrast, E7-specific CD4⁺T-cell responses were not significantly enhanced by the FL-E7 DNA vaccine. We found that the linkage of FL to E7 directly enhanced MHC class I presentation of E7 compared with wild-type E7 in transfected cells in vitro (Fig. 8). Because ballistic DNA delivery can introduce DNA directly into dermal professional APCs, the FL-E7 DNA-transfected APCs may directly enhance the presentation of E7 through MHC class I pathway to CD8⁺ T cells and contribute to the generation of E7-specific CD8⁺T-cell precursors in vivo.

Although it is not clear how the linkage of FL to E7 can directly enhance MHC class I presentation of E7, one of the possible mechanisms for the enhancement of MHC class I presentation of E7 may be related to the chaperone effect of FL. FL expressed in cells may be distributed to the ER (37). In our study, we have used fluorescence microscopic examination to investigate the distribution of E7 and FL-E7 proteins linked to GFP within transfected 293 D^b,K^b cells. In cells transfected with FL-E7-GFP, most of the FL-E7-GFP protein showed colocalization with calnexin in the ER (Fig. 1), suggesting that linkage of FL to E7 may facilitate the entry of E7 into the ER. Several studies have demonstrated that ER targeting may lead to enhanced antigen-specific MHC class I-restricted CTL activity(38, 39, 40).

Another mechanism that may contribute to enhanced E7-specific CD8⁺ T-cell immune responses in vivois the so-called “cross-priming” effect, whereby lysis of cells expressing FL-E7 antigen can release exogenous protein to be taken up and processed by other APCs via the MHC class I-restricted pathway. Our data suggested that DCs pulsed with FL-E7 fusion protein are capable of presenting E7 antigen through the MHC class I pathway in a more efficient manner than DCs pulsed with wild-type E7 protein. (Fig. 9). However, the “cross-priming” of chimeric FL-E7 probably does not play a major role in gene gun-mediated FL-E7 DNA vaccination. It has been shown that direct priming, not cross-priming, of CD8⁺ T cells by gene-transfected DCs is the key event in gene gun-mediated DNA immunization (41, 42). However, we cannot completely rule out the possibility of cross-priming, because FL-E7 might be released from other cell types,such as keratinocytes (which were also transfected by gene gun vaccination), and then enter the DCs via the cross-priming mechanism.

In this study, we did not detect a significant increase in the number of DCs or NK cells in the spleens of mice vaccinated with FL-E7 DNA vaccines (data not shown), although FL has been shown previously to significantly expand DCs (17) and NK cells (43, 44). This may be related to the low quantity of FL-E7 released in blood circulation after DNA vaccination. We were not able to detect any FL-E7 protein in the sera derived from mice vaccinated with FL-E7 DNA (data not shown). This finding also raises an issue about the source of FL-E7 protein for cross-priming. One possibility is that FL-E7 protein from the lysis of transfected keratinocytes may be taken up by Langerhans’ cells and further processed in the draining lymph nodes without involving blood circulation.

It is interesting to note that the E7 DNA vaccine in the current study had a weaker antitumor effect compared with E7 DNA vaccine using a different mammalian expression vector (21, 22). In our previous study, we used a pCMV-Neo-Bam expression vector that contains the human cytomegalovirus promoter (21, 22). The E7 DNA vaccine using this vector generated a very impressive antitumor effect with a relative absence of E7-specific CD8⁺ T-cell immune response. In the current study, we observed a weak E7-specific CD8⁺ T-cell immune response as well as a weak antitumor effect in mice vaccinated with E7 DNA using a pcDNA3 expression vector. The discrepancy in the antitumor effect generated by the same gene in difference expression vectors may be explained by the fact that different vectors may have different levels of expression of the inserted gene. Furthermore, it is now clear that bacterial DNA can contain immunostimulatory elements such as CpG islands (45, 46), which have been shown to cause simultaneous maturation and activation of murine DCs(47) and act as an adjuvant for tumor antigen immunization(48). We used pcDNA3 instead of pCMV-Neo-Bam in our current study because pCMV-Neo-Bam would likely generate a strong antitumor effect in mice vaccinated with either E7 and FL-E7 DNA,making it difficult to evaluate the correlation between E7-specific CD8⁺ T cell-mediated immune responses and the antitumor effect.

The FL-E7 DNA vaccine may raise certain safety issues that need to be addressed before it is used for widespread vaccination:

(a) There is the concern that DNA may integrate into the host genome, although it is estimated that the frequency of integration is much lower than that of spontaneous mutation and should not pose any real risk (49).

(b) The second issue concerns potential risks associated with the presence of HPV-16 E7 protein in host cells. E7 is an oncoprotein that disrupts cell cycle regulation by binding to tumor suppressor pRB protein in nuclei (50). Thus, the presence of E7 in host cells may lead to accumulation of genetic aberrations and eventual malignant transformation in the host cells. The oncogenicity of E7 can be eliminated by introducing mutations into E7 DNA so that the resulting E7 protein cannot bind with pRB (51) but still maintains most of its antigenicity.

(c) The third issue is the concern over the generation of autoimmunity that may be caused when FL leads to excessive expansion of DCs in vivo. However, we did not observe any significant increase in the number of DCs in the spleen or lymph nodes of mice vaccinated with FL or FL-E7 DNA vaccines. Furthermore, we performed pathological examination of the vital organs in all of the FL-E7-vaccinated mice, and we did not observe any significant pathology. These results indicated that FL-E7 can be used as a potent DNA vaccine with no detectable detrimental side effects.

In summary, our results indicated that fusion of the FL gene to the HPV-16 E7 gene generated potent E7-specific CD8⁺ T cell-mediated immune responses and antitumor effects against HPV-16 E7-expressing murine tumors. Our data suggest that linkage of the FL gene to an antigen gene may greatly enhance the potency of DNA vaccines and can potentially be applied to other cancer systems with known tumor-specific antigens.

We thank Drs. Keerti V. Shah, Robert J. Kurman, Elizabeth Jaffee, and Drew M. Pardoll for insightful discussions. We also thank Drs. Richard Roden and Sara Pai for critical review of the manuscript. We greatly appreciate Lee Wu for excellent statistical assistance.