Abstract
Persistent human papillomavirus (HPV)-associated benign and malignant lesions are a major cause of morbidity and mortality worldwide. Vaccination against HPV early proteins could provide an effective means of treating individuals with established infections. Recombinant vesicular stomatitis virus (VSV) vectors have been used previously to elicit strong humoral and cellular immune responses and develop prophylactic vaccines. We have shown that VSV vectors also can be used to elicit therapeutic immunity in the cottontail rabbit papillomavirus (CRPV)-rabbit model of high-risk HPV infection. In the present study, three new VSV vectors expressing the CRPV E1, E2, or E7 protein were produced and compared to the previously generated VSV-E6 vector for therapeutic efficacy. To determine whether vaccine efficacy could be augmented by simultaneous vaccination against two CRPV proteins, the four vaccines were delivered individually and in all possible pairings to rabbits 1 week after CRPV infection. Control rabbits received the recombinant wild-type VSV vector or medium only. Cumulative papilloma volumes were computed for analysis of the data. The analyses showed that VSV-based vaccination against the E1, E2, E6, or E7 protein significantly reduced papilloma volumes relative to those of the controls. Furthermore, VSV-based CRPV vaccination cured all of the papillomas in 5 of 30 rabbits. Of the individual vaccines, VSV-E7 was the most effective. The VSV-E7 vaccine alone was the most effective, as it reduced cumulative papilloma volumes by 96.9% overall, relative to those of the controls, and ultimately eliminated all of the disease in all of the vaccinees. Vaccine pairing was not, however, found to be beneficial, suggesting antigenic competition between the coexpressed CRPV proteins. These preclinical results, obtained in a physiologically relevant animal model of HPV infection, demonstrate that VSV vectors deserve serious consideration for further development as therapeutic antitumor vaccines.
Human papillomavirus (HPV) infections induce benign proliferative epithelial lesions (papillomas) at cutaneous and mucosal sites. Persistent lesions cause widespread morbidity and, when they progress to cancer, mortality (reviewed in references 1, 2, 7, 13, and 23). The new HPV virus-like particle (VLP) vaccine elicits strong HPV type-specific neutralizing antibodies and protects against subsequent infection with the corresponding HPV types (12). However, since it is unlikely to be immunotherapeutic, additional HPV vaccines are needed to ameliorate the severity of disease in the millions of people already infected with HPV (reviewed in reference 37).
A promising new approach to vaccine development uses attenuated recombinant vesicular stomatitis viruses (VSVs) as vaccines (reviewed in reference 26). Compared to other viral vectors, the VSV vector offers several advantages. It is replication competent, like the most effective human vaccines, but is not a human pathogen. It induces strong cell-mediated and humoral immune responses comparable to those elicited by wild-type VSV (reviewed in reference 30). Additionally, most human populations are seronegative for VSV (30) and therefore susceptible to VSV-based vaccination. The ability of VSV-based vaccines to protect against a subsequent challenge has been shown in a variety of animal models of human viruses, including the cottontail rabbit papillomavirus (CRPV)-rabbit model of high-risk HPV infection (6, 8, 18, 25, 27, 28, 31, 34, 35). We recently demonstrated the ability of VSV-based vaccination to induce immunotherapeutic responses in the CRPV-rabbit model. That study showed that VSV-based vaccines expressing the CRPV E6 protein significantly reduced papilloma growth and eradicated all papillomas in some rabbits (3). The present study extends those findings.
Four early papillomavirus genes are good targets for therapeutic vaccination: the E1 gene, required for DNA replication; the E2 gene, required to enhance DNA replication and to regulate the E6 and E7 promoter; and the E6 and E7 oncogenes (reviewed in references 22 and 37). All four are constitutively expressed in most high-risk papillomavirus lesions, which are premalignant. In addition, the viral E6 and E7 oncogenes are retained and constitutively expressed in all papillomavirus-associated cancers. Numerous immunization strategies targeting the E6 and/or E7 proteins have been evaluated (reviewed in reference 29), but the relative efficacy of targeting of E6 versus E7 has been evaluated in only two studies, using the CRPV-rabbit model (11). One study found no significant difference between E6 and E7 targeting to induce resistance against a subsequent CRPV challenge (22), while the other found that targeting of E7 was superior in preventing malignant progression in papilloma-bearing rabbits (11). Other studies have evaluated vaccines targeting the E1 and/or E2 proteins (4, 10, 15, 17, 36). Again, the relative efficacy of E1 versus E2 protein targeting has only been examined in two studies (21, 36). Both of these studies found no significant difference between vaccines targeting E1 versus E2 when administered prior to CRPV infection.
In the present study, VSV-based vectors individually expressing the CRPV E1, E2, and E7 proteins were generated and compared to a previously generated VSV-E6 vaccine (3) for therapeutic efficacy against CRPV-induced rabbit papillomas. Because multicomponent vaccines may elicit broader arrays of immune responses than individual vaccines, they may increase vaccine efficacy, especially among genetically heterogeneous populations such as rabbits and humans. We therefore compared the efficacy of treating rabbits with each vaccine individually with that of using each possible pairing of vaccines and analyzed the data for differences in cumulative papilloma volumes and in the frequency of papilloma regression.
MATERIALS AND METHODS
Construction of recombinant VSV vectors encoding the CRPV E1, E2, or E7 gene.
To prepare the CRPV E1 gene insert for subcloning, plasmid pcDNA3-E1K (21) was cleaved with BamHI, filled with Klenow DNA polymerase, and then cleaved with XbaI. To prepare the VSV vector, plasmid VSV-XN2 was cleaved with XhoI, filled with Klenow DNA polymerase, and then cleaved with NheI. Ligation of the two DNAs resulted in E1 VSV plasmids containing the E1 gene between the G and L genes of the VSV genome. The E2 VSV plasmid was constructed in exactly the same way with pcDNA3-E2K (21) instead of pcDNA3-E1K. The CRPV E7 gene was isolated from pShuttle-E7 with SalI and XbaI and then ligated into an XhoI-cleaved, NheI-cleaved VSV-XN2 vector to generate the E7 VSV plasmid. Plasmid pShuttle-CMV E7 was previously generated by cloning a PCR fragment containing the CRPV E7 gene flanked by an upstream KpnI site upstream and a downstream NheI site into the corresponding sites of pShuttle (14). The PCR primers were VCR1075C (5′ GCG CTG ATA TCT CGA GTC GAC CAC CAT GAT AGG CAG AAC TCC TAA GC) and VCR1359N (5′ GCG CGA GCT AGC AGA TCT ACT CAG TTA CAA CAC TCC GGG C), and the template was CRPV-pLAII (42). DNA sequencing of the resultant clones showed that the E1 and E2 genes in the E1 VSV and E2 VSV plasmids were identical to their counterparts in CRPV-pLAII, respectively. The E7 gene in the E7 VSV plasmid contained a silent transversion (TGT→TGC).
Recombinant VSV recovery.
Recombinant E1 VSV, E2 VSV, and E7 VSV were recovered as described previously (20). Briefly, BHK cells were infected with recombinant vaccinia virus vTF7-3 expressing T7 RNA polymerase. Each recombinant VSV plasmid, together with support plasmids pBS-N, pBS-P, and pBS-L under the control of T7 promoters, was transfected into vaccinia virus-infected cells. After 2 days, the supernatants were collected, filtered through a 0.2-μm-pore-diameter filter, and passaged onto fresh BHK-21 cells. The medium was collected and filtered again through a 0.1-μm-pore-diameter sterile filter. Recombinant VSVs were plaque purified and grown, and their titers were determined. Recombinants were thawed and diluted with Dulbecco modified Eagle medium to the appropriate concentration immediately before inoculation.
Production of recombinant His-tagged CRPV E1 and E2 proteins and E1- and E2-specific antisera.
His-tagged CRPV E1 and E2 protein expression vectors were constructed in pET-29a. The full-length E1 and E2 genes were first amplified by PCR from CRPV-pLAII with PCR primers E1HisC (5′ GCG CTG ACC ATG GCT GAA GGT ACA GAC CC) and E1HisN (GCG CTG ACT CGA GTA GAG ACT GAG AAG TTC C) for E1 and primers E2HisC (5′ GCG CTG ATC ATG ATC GAG GCT CTC AGC CAG CG) and E2HisN (5′ GCG CTG ACT CGA GAA GCC CAT AAA AAT TCC C) for E2. The PCR fragments were cleaved with NcoI and XhoI (for E1) or BspHI and XhoI (for E2) and ligated to the corresponding sites of pET-29a to generate His-tagged fused genes. Clones with the correct restriction endonuclease patterns were isolated from the DH5α strain of Escherichia coli and transferred to the NovaBlue strain for recombinant protein production. Isopropyl-β-d-thiogalactopyranoside (IPTG) was used to induce protein expression, and 6 M guanidinium chloride was used to lyse the cells. The lysates were individually loaded onto an Ni-nitrilotriacetic acid agarose column (QIAGEN) and eluted with 0.25 mM imidazole (Sigma). Fractions containing the highest concentrations of protein were dialyzed against Tris-buffered saline (Sigma) and pooled. Protein concentrations were determined by the Bio-Rad protein assay (Bio-Rad Laboratories).
Four 8-week-old female Hartley guinea pigs (Charles River, Wilmington, MA) were inoculated subcutaneously with 70 μg of recombinant E1 or E2 protein per immunization. Freund's complete and incomplete adjuvants were used for priming and boosting, respectively. Guinea pigs were immunized four times at approximately 3-week intervals. The guinea pig sera were tested for E1 or E2 specificity in Western blot assays with purified recombinant E1 or E2 protein, respectively.
Immunofluorescent microcopy.
BHK-21 cells were infected with each VSV recombinant or the recombinant wild-type VSV (VSV-rwt) vector. Cells were fixed in 3% paraformaldehyde for 30 min and then permeabilized with phosphate-buffered saline-glycine(1% Triton X-100) for 5 min. After washing, the cells were stained with a 1:100 dilution of primary antiserum produced in guinea pigs (anti-E1 and -E2, this study) or rabbits (anti-E6 and -E7) (38, 40). As secondary antisera, a fluorescein isothiocyanate-conjugated goat anti-guinea pig immunoglobulin G (IgG) antiserum (Antibodies Inc.) was diluted 1:200 and an Alexa Fluor 488-conjugated goat anti-rabbit IgG (HPV+L) antiserum (Invitrogen) was diluted 1:250. Photographs were taken on a Nikon Microphot FX microscope equipped with a 40× Planapochromat objective, epifluorescence, and a SPOT digital camera.
Western blotting.
BHK-21 cells were mock infected or infected with VSV-rwt, VSV-E1, or VSV-E2 as described above. Cells were lysed in Laemmli buffer containing 5% β-mercaptoethanol and 2% sodium dodecyl sulfate (SDS), and the DNA was sheared with a 27-gauge needle. Proteins were separated by 12% SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to an Immun-Blot polyvinylidene difluoride membrane (Bio-Rad) in a Mini Trans-Blot electrophoretic transfer cell (Bio-Rad). The blot was probed with guinea pig anti-E1 or anti-E2 antiserum at a dilution of 1:1,500, followed by horseradish peroxidase-conjugated anti-guinea pig serum (Kirkegaard & Perry Laboratories, Inc.) at a dilution of 1:5,000. The blot was developed with Western Lighting Chemiluminescence Reagent Plus (Perkin-Elmer Life and Analytical Sciences, Inc.) and exposed to Kodak BioMax MR film (Kodak).
Metabolic labeling, immunoprecipitation, and SDS-PAGE.
BHK cells were mock infected or infected with VSV-rwt, VSV-E6, or VSV-E7 as described previously (3). After 5 h, the cells were incubated in Dulbecco modified Eagle medium containing 100 μCi of [35S]cysteine for 1 h and then lysed as described previously (3). The CRPV E6 and E7 proteins were immunoprecipitated from the lysates with rabbit anti-E6 (40) and anti-E7 (38) antisera at a dilution of 1:100, and the proteins were separated by 15% SDS-PAGE as described previously (3).
CRPV infection of rabbits.
Two-kilogram female New Zealand White Pasteurella-free rabbits (Charles River, Wilmington, MA) were maintained in the animal facilities at the Yale University School of Medicine. All experiments were performed in accordance with procedures approved by the Yale Institutional Animal Care and Use Committee. Thirty-six rabbits were infected with CRPV at nine cutaneous sites on the right flank as described previously (40). Briefly, three sites were infected with 10 μl of a 1:30, 1:150, or 1:750 dilution of the K216 stock of CRPV (39).
VSV vaccination.
Immediately prior to vaccination, rabbits were anesthetized by intramuscular injection of Acepromazine (35 mg/kg) and fur was clipped from the shoulder. The shoulder was then inoculated intradermally with a dose of 4 × 107 PFU VSV in 0.4 ml cell culture medium or with 0.4 ml cell culture medium alone at approximately 20 adjacent sites on the shoulder. When pairs of vaccines were administered, they were mixed prior to inoculation and a half dose of each component was used.
Collection of clinical data.
Rabbits were examined by visual inspection and palpation after clipping the fur 18, 21,26, 32, 39, 46, 54, 62, 68, 75, and 82 days after CRPV infection. At each examination, the numbers, locations, and dimensions of papillomas were recorded in millimeters (length, width, height). Papilloma volumes were calculated in cubic millimeters by using the formula for an irregular sphere, the shape of papillomas, i.e., 4/3 × π × length/2 × width/2 × height/2.
VSV-neutralizing antibody titers.
Neutralization assays were performed as previously described to detect antibodies that neutralize VSV (Indiana serotype) by binding to the VSV G protein (24).
ELISAs.
Enzyme-linked immunosorbent assay (ELISA) plates were coated with 200 ng of recombinant CRPV protein E1, E2 (this paper), E6 (40), or E7 (38). Positive control antisera were described previously (4). Negative control sera were from preimmune rabbits. Test sera were collected 10 weeks after vaccination. The test and negative control sera were assayed at 1:40 to 1:320, and the positive control sera were assayed at 1:100 to 1:12,800 (for E1 and E2) or 1:100 to 1:6,400 (for E6 and E7), by using serial twofold dilutions. Serum samples with undetectable responses at the 1:40 dilution were designated negative.
Statistical analyses.
The efficacy of each treatment was summarized by using cumulative papilloma volumes at each CRPV infection site as the primary outcome measurement. For sites where no papilloma was detected, a volume of 0.00065 mm3 was used to enable analysis of the log volumes. This is the theoretical volume of the smallest papilloma recorded (5 by 5 by 5 mm), divided by 105, 10 times the minimum number of cells estimated to allow clinical detection. The cumulative volume was calculated by using the trapezoidal rule for the area under the curve generated by plotting the volume at each site over time. Analyses of volume effects used log volumes because of the highly skewed values but are presented in the original linear scale. The distribution of the cumulative papilloma volumes was heteroscedastic, so the conventional parametric analysis of variance (ANOVA) was not applicable. Instead, the volume data were analyzed by using rank-score tests in factorial designs following the nonparametric approach of Brunner et al. (5). The raw data were ranked and modeled with the PROC MIXED procedure of SAS version 9.1 with an unconstrained within-animal covariance structure.
The effects of the CRPV dose on papilloma frequency and regression frequency were analyzed by chi-squared tests and logistic regression. Complete papilloma regression was defined as the disappearance of all lesions, and partial regression was defined as the disappearance of at least one papilloma but not all of the papillomas. The effect of the dose on the time to first detection of a papilloma (by visualization and palpation) was analyzed by both unpaired Student's t tests for two samples of unequal variance and logistic regression models.
The correlation between VSV-specific neutralizing antibody titers and cumulative papilloma volumes was described by the Pearson correlation coefficient. Differences in the frequency of papilloma regression, defined as the complete disappearance of a papilloma, were analyzed by the Cochran Armitage trend test. The responsiveness of rabbits to vaccination according to the dose of CRPV used for infection was tested by a latent parametric model.
RESULTS
Construction and characterization of VSV recombinants expressing CRPV early proteins E1, E2, E6, and E7.
We have previously demonstrated that replication-competent VSV can be genetically engineered to express the CRPV E6 protein and that this vector can generate therapeutic immunity in CRPV-infected rabbits (21). To assess the immunotherapeutic potential of VSV-based vectors expressing other CRPV early proteins, three recombinant plasmids encoding the full-length E1, E2, or E7 protein of CRPV were constructed. Each CRPV gene was flanked by appropriate VSV transcription start and stop sites and inserted between the G and L genes of the VSV genome (Fig. 1). Recombinants VSV-E1, VSV-E2, and VSV-E7 were recovered as described previously (20). These three recombinants, VSV-E6 (VSV-E6.5 in reference 3), and VSV-rwt were used in this study.
FIG. 1.
The VSV cloning vector. The genome of the parental VSV-rwt vector is diagrammed in a 3′-to-5′ orientation on the negative-stranded viral RNA genome. Letters refer to the VSV nucleocapsid (N), phosphoprotein (P), matrix (M), glycoprotein (G), and RNA-dependent RNA polymerase (L) genes. The CRPV genes were inserted into position 5 of the VSV genome, between the G and L genes, and expressed by duplication of the VSV start and stop signals.
CRPV protein expression was assessed by immunofluorescent staining of VSV-infected BHK cells with guinea pig antisera that we produced against recombinant His-tagged CRPV E1 and E2 proteins (Fig. 2B) and rabbit antisera previously produced against CRPV E6 and E7 (38, 40). As shown in Fig. 2A, each VSV recombinant specifically expressed the CRPV protein it encoded. The E1 and E2 proteins were expressed in the cytoplasm, and some VSV-E2-infected cells additionally showed strong perinuclear staining. The E6 and E7 proteins were primarily nuclear, and the level of E7 expression was more variable with especially bright staining in some cells. E1, E2, E6, and E7 protein expression in VSV-infected cells was confirmed by Western blotting (Fig. 2C) or immunoprecipitation (Fig. 2D). Each recombinant VSV vector expressed a protein of the appropriate size for the corresponding CRPV protein, whereas mock-infected and vector-infected control cells were negative.
FIG. 2.
Detection of the VSV-encoded E1, E2, E6, and E7 proteins. (A) BHK cells infected with VSV-E1, VSV-E2, VSV-E6, VSV-E7 (top row), or VSV-rwt (bottom row). Cells in both rows were stained with a primary antiserum corresponding to the protein encoded by the vector and a secondary antiserum against guinea pig IgG (to detect E1 and E2) or rabbit IgG (to detect E6 and E7). The exposure times for the controls were identical to the exposure times for the experimental cells. (B) SDS-polyacrylamide gel showing the purified recombinant His-tagged E1 and E2 proteins (lanes 1 and 2, respectively) used to generate the guinea pig antisera to E1 and E2. (C and D) BHK cell lysates infected with VSV-E1 (C1), VSV-E2 (C2), VSV-E6 (D1), and VSV-E7 (D2) and processed for Western blotting (C1 and C2) or immunoprecipitation (D1 and D2). The lanes in panels C and D contain lysates of BHK cells that were mock infected (lane 1) or infected with VSV-rwt (lane 2) or a VSV-CRPV recombinant virus (lane 3). In each panel, the band corresponding to the CRPV protein is labeled with its name.
Design of the immunotherapy experiment.
Following the generation of replication-competent VSV-based CRPV early-gene vectors and demonstration of recombinant CRPV protein expression, an immunotherapeutic experiment was performed to test the efficacy of the VSV-based vaccines. In this experiment, 36 rabbits were infected with CRPV at each of three sites with 1:30, 1:150, and 1:750 dilutions of virus, i.e., high, moderate, and low doses (nine sites per rabbit). One week later, 12 randomly formed groups of three rabbits were treated intradermally on the contralateral flank. Four groups received one VSV-CRPV vaccine, and six groups received each possible pair of vaccines. Two control groups received the VSV-rwt vector or medium alone. As shown in Table 1, this design provided 12 rabbits to estimate the effects of each vaccine; e.g., rabbits in groups A, B, C, and D were used to analyze the effects of VSV-E1. All VSV-inoculated rabbits received a dose of 4 × 107 PFU of VSV; i.e., rabbits inoculated with a mixture of two vectors received a half dose of each one. Half doses were used in the pairs so that differences between treatments, if they occurred, could be attributed to the nature of the treatment (one vaccine component versus two vaccine components) rather than the dose of vaccine. Papilloma data were recorded approximately weekly until 11 weeks after vaccination, when the rabbits were euthanized. The only rabbit whose papillomas were actively regressing at the 11-week time point was held 7 weeks longer to determine whether further regression would occur.
TABLE 1.
Experimental design used in this study
| Vaccine component | Groupa also receiving following vaccine component:
|
||||
|---|---|---|---|---|---|
| VSV-E1 | VSV-E2 | VSV-E6 | VSV-E7 | None | |
| VSV-E1 | A | B | C | D | |
| VSV-E2 | E | F | G | ||
| VSV-E6 | H | I | |||
| VSV-E7 | J | ||||
| VSV-rwt | K | ||||
| Medium | L | ||||
Rabbits vaccinated with a single component (groups A, E, H, J, and K) received 4 × 107 PFU. Rabbits vaccinated with two components (groups B, C, D, F, G, and I) received 2 × 107 PFU of each one.
Papilloma formation and effects of CRPV dose.
Analysis of the frequency of papilloma formation and the time a papilloma was first detected showed that vaccination did not affect these variables. At the high and moderate doses of CRPV, all of the sites in all of the rabbits formed papillomas, indicating that the high-dose sites received at least five times the minimum papilloma-inducing dose of virus (Table 2). At the low-dose sites, 85% formed papillomas, which, while still a high frequency, was significantly lower compared to the moderate-dose sites (P = 0.00003). The time points when a papilloma was first detected ranged from <19.0 to 29.1 days and again were highly dependent on the dose of CRPV administered (P = 0.0001 for high versus moderate or moderate versus low doses) (Table 2). Finally, papilloma volumes were significantly affected by the CRPV dose; they were 1.8-fold larger at the high- versus moderate-dose sites (P = 0.012) and 2.1-fold larger at the moderate- versus low-dose sites (P = 0.004) (Table 2). Papilloma volumes were also strongly affected by vaccination, as presented below.
TABLE 2.
Effect of CRPV dose on papilloma outcome
| CRPV dilution | Papilloma frequencya | First day of detectionb | Maximum vol/sitec |
|---|---|---|---|
| 1:30 | 108/108 | <19.0d ± 0.2 | 209 ± 17 |
| 1:150 | 108/108 | 23.0 ± 0.6e | 118 ± 12e |
| 1:750 | 92/108e | 29.1 ± 0.8e | 57 ± 7e |
Maximum number of positive sites/all sites.
The first examination day when a papilloma was detected clinically.
Volume, in cubic millimeters, for all rabbits. For volumes according to treatment, see Table 4.
Less than 19 days because most sites were already positive on day 18, the first day rabbits were examined.
P ≤ 0.00003 for 1:30 versus 1:150 or 1:150 versus 1:750.
Immunogenicity of VSV-based CRPV vaccines.
Since recombinant viral vaccines elicit immune responses to vector antigens, as well as recombinant targets, titers of neutralizing antibody to the VSV vector were assessed as an initial measurement of vaccine immunogenicity. None of the controls inoculated with medium alone developed detectable VSV-neutralizing antibody. In contrast, all of the rabbits inoculated with any VSV vector(s) developed VSV-specific neutralizing antibodies with mean titers ranging from 1:640 to 1:1,813 in the individual groups, with no significant differences between any two groups or between the VSV-rwt group and the CRPV VSV-vaccinated groups (Table 3). This level of variation is normal in genetically heterogeneous populations such as rabbits. Further analysis showed a lack of correlation between VSV-specific neutralizing antibody titers in the vaccinated rabbits and papilloma volumes (correlation coefficient = −0.10933, P = 0.57).
TABLE 3.
VSV-neutralizing antibody titers
| Vaccine component | Result obtained by administration of following vaccine component:
|
||||
|---|---|---|---|---|---|
| VSV-E1 | VSV-E2 | VSV-E6 | VSV-E7 | None | |
| VSV-E1 | 1,707 ± 427a | 480 ± 160 | 747 ± 160 | 1,814 ± 747 | |
| VSV-E2 | 1,280 ± 0 | 1,280 ± 0 | 1,813 ± 747 | ||
| VSV-E6 | 1,707 ± 427 | 533 ± 107 | |||
| VSV-E7 | 1,280 ± 0 | ||||
| VSV-rwt | 640 ± 320 | ||||
| Medium | Negativeb | ||||
Reciprocal of highest dilution with a positive ELISA result (mean ± SEM).
Not detected at the lowest dilution assayed (1:40).
Effects of vaccination on cumulative papilloma volumes.
The antitumor efficacy of VSV-based CRPV early-gene vaccination was evaluated by using the cumulative papilloma volumes at each infection site. As cumulative volumes measure the total papilloma burden carried by the rabbits throughout the experiment, it is a stringent criterion for judging efficacy. The distribution of cumulative papilloma volumes in each treatment group and at each CRPV dose show that the raw data were not normally distributed and skewed (Fig. 3), which precluded analysis by the conventional parametric ANOVA. Instead, the nonparametric approach of Brunner et al. (5) was used for data analysis.
FIG. 3.
Nonnormal distribution of the raw data. The plot shows cumulative papilloma volumes according to the dose of CRPV used for infection and the VSV recombinants used for vaccination. Each box contains 50% of the data. The vertical lines show the top (above the box) and bottom (below the box) 25% of the data. The horizontal line within each box shows the median value. Med, medium; Vec, vector.
The therapeutic efficacy of vaccination on cumulative papilloma volumes was first analyzed according to the individual CRPV target proteins. The effects of targeting E1, for example, were analyzed by considering its contributions to groups A, B, C, and D while taking into account the effects of the other vaccine components and their corresponding pairing effects (Table 1). Vaccination against each CRPV protein significantly reduced papilloma volumes (Tables 4 and 5), and relative to the controls, the strongest effects were elicited by targeting the E7 protein (P < 0.0001). However, compared to those of the other vaccines, the effects of VSV-E7 were not significantly greater.
TABLE 4.
Cumulative papilloma volumes in rabbits treated with VSV-based CRPV vaccine(s)
| CRPV dilution and vaccine component | Result obtained by administration of following vaccine component:
|
||||
|---|---|---|---|---|---|
| VSV-E1 | VSV-E2 | VSV-E6 | VSV-E7 | None | |
| 1:30 | |||||
| VSV-E1 | 6,271 ± 1,298a | 7,926 ± 1,258 | 4,683 ± 670 | 3,990 ± 370 | |
| VSV-E2 | 2,937 ± 751 | 3,857 ± 1,762 | 4,250 ± 1,205 | ||
| VSV-E6 | 6,414 ± 331 | 4,346 ± 1,184 | |||
| VSV-E7 | 553 ± 285 | ||||
| VSV-rwt | 13,355 ± 2,241 | ||||
| Medium | 9,084 ± 879 | ||||
| 1:150 | |||||
| VSV-E1 | 1,853 ± 339 | 4,236 ± 702 | 2,096 ± 325 | 2,433 ± 422 | |
| VSV-E2 | 1,223 ± 340 | 1,804 ± 865 | 2005 ± 646 | ||
| VSV-E6 | 3,458 ± 570 | 1,629 ± 671 | |||
| VSV-E7 | 49 ± 21 | ||||
| VSV-rwt | 7,288 ± 1,660 | ||||
| Medium | 4,828 ± 470 | ||||
| 1:750 | |||||
| VSV-E1 | 1,099 ± 364 | 1,540 ± 411 | 1,187 ± 271 | 824 ± 234 | |
| VSV-E2 | 406 ± 235 | 655 ± 333 | 549 ± 178 | ||
| VSV-E6 | 1,045 ± 153 | 533 ± 231 | |||
| VSV-E7 | 1.6 ± 1.0 | ||||
| VSV-rwt | 2,732 ± 979 | ||||
| Medium | 2,106 ± 539 | ||||
Volume, in cubic millimeters (mean ± SEM). Values are descriptive; they were not used to calculate statistical significance.
TABLE 5.
Analysis of treatment effects on cumulative papilloma volumes
| Effect and vaccine(s) | ATSa | P value |
|---|---|---|
| Vaccinationb | ||
| VSV-E1 | 3.97 | 0.0462 |
| VSV-E2 | 7.35 | 0.0067 |
| VSV-E6 | 5.29 | 0.0215 |
| VSV-E7 | 80.3 | <0.0001 |
| Interactionc | ||
| VSV-E1-VSV-E2 | 7.84 | 0.0051 |
| VSV-E1-VSV-E6 | 0.78 | 0.38d |
| VSV-E1-VSV-E7 | 22.32 | <0.0001 |
| VSV-E2-VSV-E6 | 0.24 | 0.63d |
| VSV-E2-VSV-E7 | 12.41 | 0.0004 |
| VSV-E6-VSV-E7 | 7.41 | 0.0065 |
ATS, ANOVA-type statistic.
The effects of each vaccine were analyzed by considering its contributions to the four groups receiving it (Table 1) while taking into account the effects of the other vaccines and their corresponding pairing effects.
The effects of each pair of vaccines were analyzed by considering the pair's effects in the three rabbits receiving it (Table 1) after taking away the effects of the vaccines in the pair.
No statistically significant difference.
Next, we analyzed the effects of pairing, in a single inoculum, two vaccines encoding different CRPV proteins, which we hypothesized would be therapeutically beneficial. The effects of pairing VSV-E1 and VSV-E2, for example, were analyzed by considering the effects in the three rabbits in group B (Table 1) after taking away the effects of VSV-E1 and VSV-E2. Contrary to expectations, we did not detect positive synergy for any pair relative to its more effective component (Table 5). For the two pairs containing VSV-E6 plus either VSV-E1 or VSV-E2, the effects may have been additive but they were not synergistic. For the other four pairs, including all three containing VSV-E7, pairing was significantly detrimental (negative synergy) (P ≤ 0.01) (Table 5). Potential reasons for this outcome are discussed below.
The data were further analyzed to determine whether VSV-based vaccination was more effective against smaller versus larger papillomas, i.e., induced by lower versus higher doses of CRPV. Such a possibility was suggested by the data shown in Table 4. However, when analyzed statistically, the difference in reduction percentages (mean volume of the CRPV VSV vaccinees/mean volume of the controls) according to the CRPV dose administered was not significant (P = 0.28).
Effects of vaccination on papilloma regression.
The ultimate goal of therapeutic vaccination is a complete clinical response, which for papillomavirus-induced disease means the elimination of all viral lesions (complete regression). Within the time frame of the main experiment, VSV-based CRPV vaccination induced the regression of 36/255 papillomas in the vaccinees versus 0/53 in the controls (P = 0.001). Most regressions occurred in four rabbits that showed complete papilloma regression, i.e., one treated with VSV-E2 alone, one treated with the VSV-E2-VSV-E6 pair, and two treated with VSV-E7 alone (Table 6). Many fewer regressions occurred in the four rabbits with partial regression (Table 6). Further analysis showed that regression occurred significantly earlier among papillomas induced by the low versus higher doses of CRPV, i.e., 34.5 ± 2.8 versus 47.2 ± 3.3 days after vaccination, respectively (mean ± standard error of the mean [SEM]) (P = 0.020). Additionally, among the partial regressors, six of the seven papillomas that regressed were induced by the low CRPV dose (P < 0.001). In summary, papillomas induced by the low CRPV dose regressed more rapidly than other papillomas among all regressors and more frequently among partial regressors.
TABLE 6.
Frequency of papilloma regression 11 weeks after vaccination
| Vaccine component | Result obtained by administration of following vaccine component:
|
||||
|---|---|---|---|---|---|
| VSV-E1 | VSV-E2 | VSV-E6 | VSV-E7 | None | |
| VSV-E1 | 1/27a | 0/27 | 0/26 | 0/26 | |
| VSV-E2 | 6/22b | 10/25c | 2/27d | ||
| VSV-E6 | 0/27 | 0/25 | |||
| VSV-E7 | 17/23e | ||||
| VSV-rwt | 0/27 | ||||
| Medium | 0/26 | ||||
In one rabbit, one of nine papillomas regressed.
In one rabbit, six of six papillomas regressed.
In one rabbit, nine of nine papillomas regressed; in another rabbit, one of seven papillomas regressed.
In one rabbit, two of nine papillomas regressed.
In each of two rabbits, seven of seven papillomas regressed; in the third rabbit, three of nine papillomas regressed. In the partial regressor, the remaining six papillomas regressed 7 weeks later.
The frequency of complete or partial regression was significantly greater among rabbits treated with VSV-E7 alone than among the controls (P = 0.0046). Furthermore, papillomas in the VSV-E7 vaccinee with partial regression were actively regressing at the 11-week time point. She was therefore held longer, and complete regression occurred 7 weeks later. Although only this rabbit was held longer, it is unlikely that any others would have undergone complete regression in the same time frame because none of them showed signs of regression at the 11-week time point. Thus, the VSV-E7 vaccine not only induced the greatest reductions in cumulative papilloma volumes (Table 4) but also eliminated all disease in three of three vaccinees.
Photographic documentation of the time course of papilloma growth versus regression shows two rabbits with multiple papillomas 25 days after vaccination (Fig. 4A). In one (control) rabbit, papillomas continued to grow (top), whereas in the other (regressor) rabbit, papillomas diminished in size on day 33 and disappeared on day 48 (bottom). A plot of the kinetics of papilloma growth versus regression shows that papillomas in all five regressors began diminishing in volume 4 weeks after treatment, i.e., relatively rapidly after treatment (Fig. 5). Regression was complete after 6 weeks in the VSV-E2-VSV-E6 vaccinee, 9 weeks in the VSV-E2-alone vaccinee, and 7.8 or more than 11 weeks in the three VSV-E7 vaccinees. The rabbit that developed the largest papillomas also was the one that took the longest for complete regression.
FIG. 4.
Clinical aspects of papilloma regression. (A) Time course of papilloma growth in a control rabbit (top) versus papilloma regression in a regressor rabbit (bottom). The photographs were taken 25, 33, and 48 days after vaccination. The sites to the left, middle, and right of each photograph were infected with 1:750, 1:150, and 1:30 dilutions of CRPV, respectively. Rabbit fur was clipped prior to photography. (B) Clinical status of one control and the five complete-regressor rabbits, labeled with the treatment they received. The photographs were taken at the end of the main experiment. The last rabbit (lower right) shows papillomas during active regression; the arrow marks the only papilloma of substantive size involuting at the base.
FIG. 5.
Kinetics of papilloma growth and regression in the regressors. Mean papilloma volumes per rabbit are shown for a representative control rabbit (Control) and the five complete regressor rabbits treated with VSV-E7 alone (E7), VSV-E2 plus VSV-E6 (E2+E6), or VSV-E2 alone (E2). Panels A and B show the same data on different scales.
The clinical statuses of the five complete regressors versus a representative control rabbit at the end of the main experiment are shown in Fig. 5B. The regression sites in the first four regressors were virtually indistinguishable from normal skin, except for a large amount of hair at the high-dose sites of one regressor (Fig. 5B, bottom, center). As regression involves the replacement of hairless papillomas with hair-bearing skin, ample local hair growth is a typical last sign of prior cutaneous infection. In the fifth regressor, papillomas at the three low-dose sites had already regressed and those at the remaining sites had diminished in volume (Fig. 5B, bottom, right). In fact, the only remaining papilloma of substantive size was involuting at the base (arrow in Fig. 5B, bottom, right), indicating that it would soon fall off, as others had already done. As stated above, 7 weeks later all signs of disease in this rabbit disappeared. Thus, treatment with VSV-E7 alone induced complete regression in all three vaccinees.
Effects of vaccination on CRPV-specific humoral immunity.
ELISAs were performed to determine whether the VSV-based vaccination elicited antibody responses to the CRPV E1, E2, E6, and/or E7 proteins and, if so, whether they correlated with papilloma volumes. E1-specific antibodies developed in 6/12 VSV-E1-immunized versus 4/24 other rabbits (P = 0.035), with a mean reciprocal titer of 192 ± 30 for all seropositive rabbits (mean ± SEM). E2-specific antibodies developed in 7/12 VSV-E2-immunized rabbits versus 5/24 other rabbits (P = 0.024), with a mean reciprocal titer of 73 ± 13 for all seropositive rabbits. These results show that CRPV infection itself was sufficient to elicit antibody responses to E1 and/or E2 in some rabbits, although the frequency of responding rabbits was significantly increased by VSV-E1 or VSV-E2 vaccination, respectively. The results also show that the titers of antibodies to E1 and E2 were universally quite low. Further analysis showed no correlation between E1/E2-specific humoral immunity and papilloma volumes. Antibody responses to the E6 and E7 proteins were virtually absent, except in one E7 vaccinee and one control, which each developed a 1:80 titer of antibody to E7. Taken together, these results indicate that therapeutic immunity was not mediated by antibody responses to any of the vaccine targets.
DISCUSSION
VSV-based vaccines encoding the early papillomavirus proteins represent a promising new strategy to elicit therapeutic immunity for the treatment of papillomavirus-associated lesions. Previously, we showed that VSV-based vaccines targeting the CRPV E6 protein induced therapeutic immunity in the CRPV-rabbit model. The E7 protein is the other viral target for the immunotherapy of papillomavirus-associated malignant lesions, and the E1, E2, E6, and E7 proteins are all suitable targets for the immunotherapy of all premalignant lesions. We undertook the present study to compare the therapeutic efficacy of VSV-based vaccines targeting the papillomavirus E1, E2, E6, or E7 protein in the CRPV-rabbit model. Each vector expressed the CRPV gene from the same position in the VSV genome (between the G and L genes) to control for the levels of transcription.
Each of the VSV-based vaccines targeting the CRPV E1, E2, E6, or E7 protein significantly reduced the cumulative papilloma volumes in rabbits immunized 1 week after CRPV infection, relative to inoculation with the VSV-rwt vector or cell culture medium alone. One vector (VSV-E6) was previously shown to reduce papilloma growth by 54% relative to that in control rabbits, and it performed similarly in the present study, reducing papilloma volumes by 55% overall (Table 4). VSV-based CRPV vaccination also cured all papillomas in 5 of 30 rabbits, i.e., 4 within 9 weeks of vaccination and 1 after 18 weeks. We cannot rule out the possibility of additional cures in other rabbits had they been held longer than 18 weeks. Smaller papillomas tended to undergo larger percent reductions in volume than larger papillomas, but this effect did not reach statistical significance. On the other hand, smaller papillomas regressed significantly more rapidly and more frequently than larger papillomas. These results suggest that the ultimate success of a therapeutic HPV vaccine will partly depend on the size of the lesions being treated.
By virtue of inducing immune responses to a broader array of antigens, multicomponent vaccines have the potential to be more effective than single-component vaccines. This potential is not always realized, however, and in a previous study we found that prophylactic vaccination of rabbits with a pair of DNA vectors individually encoding the CRPV E6 or E7 protein was less effective than vaccination with either vector alone (21). In the present study, we did not detect a synergistic benefit of pairing the VSV-E6 vaccine with either the VSV-E1 or the VSV-E2 vaccine, although the effects may have been additive. In contrast, we detected significantly negative synergy for four of the six vaccine pairs, including all pairs containing VSV-E7.We do not know why the paired vaccines did not show positive synergy. Possibly, the levels of CRPV protein-specific immunity induced by the two half doses of vaccine were below a threshold required for optimal therapeutic efficacy, in which case increased dosing might prove beneficial. Alternatively or in addition, reduced efficacy may have resulted from antigenic competition among the antigens of coexpressed CRPV proteins. Several mechanisms for antigenic competition have been described, including competition for antigen uptake by antigen-presenting cells (APCs), competition of APCs for (in this case) T-helper type 1-related cytokines, and competition of T cells, including T regulatory cells, for physical access to limited numbers of APCs (16, 19, 41). Future studies could determine whether superior efficacy could be achieved by vaccinating different sites with different VSV-based CRPV vectors to avoid antigenic competition, as in other studies (33). More importantly, the results show that multicomponent vaccines are not necessarily superior to single-component vaccines and that the two strategies should be directly compared prior to proceeding with a combined vaccine.
Three findings suggest that the VSV-E7 vaccine was superior to the other vaccines; i.e., it induced by far the greatest reductions in papilloma volumes; its efficacy was significantly impaired when it was combined with VSV-E1, VSV-E2, or VSV-E6; and it eliminated all disease in three of three vaccinees when used alone. Nevertheless, the effects of VSV-E7 on papilloma volumes were not significantly different from the effects of the other vaccines. Additionally, only three rabbits were treated with each vaccine pair, making analysis of the pairing effects more sensitive to individual differences among the rabbits than analysis of the effects of VSV-E1, VSV-E2, VSV-E6, and VSV-E7. Indeed, genetic polymorphisms in the major histocompatibility complex locus affect the presentation of various proteins to T cells and certain major histocompatibility complex class II alleles have been linked to spontaneous papilloma regression versus progression in rabbits (9). Other genetic factors must also have controlled immune responses, and yet others must have controlled the relative susceptibility of different rabbits to CRPV-induced disease. Understanding the role of the cellular immune response in tumor clearance will require the development of quantitative assays for cellular immunity in the rabbit model.
Taken together, our results provide a solid foundation for the use of VSV-based vectors as therapeutic vaccines. The ability to reduce papilloma volumes is of considerable value because smaller lesions will be easier to treat by complementary modalities. They also will progress to carcinoma less frequently and less rapidly than larger lesions (32; unpublished data). Finally, the ability of VSV-based vaccination to induce complete papilloma regression in a physiologically relevant animal model of tumorigenesis (this study and reference 3) demonstrates that VSV-based vectors deserve serious consideration for the further development of therapeutic anti-tumor vaccines against other tumors, as well as those associated with HPV infection.
Acknowledgments
This work was supported by grants from the American Cancer Society (TURG MBC-100103) and the Natural Cancer Institute (RO1-98355) to J.L.B.
Footnotes
Published ahead of print on 28 March 2007.
REFERENCES
- 1.Aaltonen, L. M., H. Rihkanen, and A. Vaheri. 2002. Human papillomavirus in larynx. Laryngoscope 112:700-707. [DOI] [PubMed] [Google Scholar]
- 2.Ahmed, A. M., V. Madkan, and S. K. Tyring. 2006. Human papillomaviruses and genital disease. Dermatol. Clin. 24:157-165. [DOI] [PubMed] [Google Scholar]
- 3.Brandsma, J. L., M. Shlyankevich, L. Buonocore, A. Roberts, S. M. Becker, and J. K. Rose. 2007. Therapeutic efficacy of vesicular stomatitis virus-based E6 vaccination in rabbits. Vaccine 25:751-762. [DOI] [PubMed] [Google Scholar]
- 4.Brandsma, J. L., M. Shlyankevich, L. Zhang, M. D. Slade, E. C. Goodwin, W. Peh, and A. B. Deisseroth. 2004. Vaccination of rabbits with an adenovirus vector expressing the papillomavirus E2 protein leads to clearance of papillomas and infection. J. Virol. 78:116-123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Brunner, E., U. Munzel, and M. L. Puri. 1999. Rank-score tests in factorial designs with repeated measures. J. Multivar. Anal. 70:286-317. [Google Scholar]
- 6.Foley, H. D., J. P. McGettigan, C. A. Siler, B. Dietzschold, and M. J. Schnell. 2000. A recombinant rabies virus expressing vesicular stomatitis virus glycoprotein fails to protect against rabies virus infection. Proc. Natl. Acad. Sci. USA 97:14680-14685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gillison, M. L., and K. V. Shah. 2003. Chapter 9: role of mucosal human papillomavirus in nongenital cancers. J. Natl. Cancer Inst. Monogr. 2003:57-65. [DOI] [PubMed] [Google Scholar]
- 8.Haglund, K., I. Leiner, K. Kerksiek, L. Buonocore, E. Pamer, and J. K. Rose. 2002. High-level primary CD8+ T-cell response to human immunodeficiency virus type 1 Gag and Env generated by vaccination with recombinant vesicular stomatitis. viruses. J. Virol. 76:2730-2738. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Han, R., F. Breitburd, P. N. Marche, and G. Orth. 1992. Linkage of regression and malignant conversion of rabbit viral papillomas to MHC class II genes. Nature 356:66-68. [DOI] [PubMed] [Google Scholar]
- 10.Han, R., N. M. Cladel, C. A. Reed, X. Peng, and N. D. Christensen. 1999. Protection of rabbits from viral challenge by gene gun-based intracutaneous vaccination with a combination of cottontail rabbit papillomavirus E1, E2, E6, and E7 genes. J. Virol. 73:7039-7043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Han, R., X. Peng, C. A. Reed, N. M. Cladel, L. R. Budgeon, M. D. Pickel, and N. D. Christensen. 2002. Gene gun-mediated intracutaneous vaccination with papillomavirus E7 gene delays cancer development of papillomavirus-induced skin papillomas on rabbits. Cancer Detect. Prev. 26:458-467. [DOI] [PubMed] [Google Scholar]
- 12.Harper, D. M., E. L. Franco, C. M. Wheeler, A.-B. Moscicki, B. Romanowski, C. M. Roteli-Martins, D. Jenkins, A. Schuind, S. A. Costa Clemens, G. Dubin, and the HPV Vaccine Study Group. 2006. Sustained efficacy up to 4.5 years of a bivalent L1 virus-like particle vaccine against human papillomavirus types 16 and 18: follow-up from a randomised control trial. Lancet 367:1247-1255. [DOI] [PubMed] [Google Scholar]
- 13.Harwood, C. A., and C. M. Proby. 2002. Human papillomaviruses and non-melanoma skin cancer. Curr. Opin. Infect. Dis. 15:101-114. [DOI] [PubMed] [Google Scholar]
- 14.He, T. C., S. Zhou, L. T. da Costa, J. Yu, K. W. Kinzler, and B. Vogelstein. 1998. A simplified system for generating recombinant adenoviruses. Proc. Natl. Acad. Sci. USA 95:2509-2514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Jensen, E. R., R. Selvakumar, H. Shen, R. Ahmed, F. O. Wettstein, and J. F. Miller. 1997. Recombinant Listeria monocytogenes vaccination eliminates papillomavirus-induced tumors and prevents papilloma formation from viral DNA. J. Virol. 71:8467-8474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Jin, X., M. A. Demoitie, S. M. Donahoe, G. S. Ogg, S. Bonhoeffer, W. M. Kakimoto, G. Gillespie, P. A. Moss, W. Dyer, M. G. Kurilla, S. R. Riddell, J. Downie, J. S. Sullivan, A. J. McMichael, C. Workman, and D. F. Nixon. 2000. High frequency of cytomegalovirus-specific cytotoxic T-effector cells in HLA-A*0201-positive subjects during multiple viral coinfections. J. Infect. Dis. 181:165-175. [DOI] [PubMed] [Google Scholar]
- 17.Johnston, K. B., J. M. Monteiro, L. D. Schultz, L. Chen, F. Wang, V. A. Ausensi, E. C. Dell, E. B. Santos, R. A. Moore, T. J. Palker, M. A. Stanley, and K. U. Jansen. 2005. Protection of beagle dogs from mucosal challenge with canine oral papillomavirus by immunization with recombinant adenoviruses expressing codon-optimized early genes. Virology 336:208-218. [DOI] [PubMed] [Google Scholar]
- 18.Kahn, J. S., A. Roberts, C. Weibel, L. Buonocore, and J. K. Rose. 2001. Replication-competent or attenuated, nonpropagating vesicular stomatitis viruses expressing respiratory syncytial virus (RSV) antigens protect mice against RSV challenge. J. Virol. 75:11079-11087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kedl, R. M., W. A. Rees, D. A. Hildeman, B. Schaefer, T. Mitchell, J. Kappler, and P. Marrack. 2000. T cells compete for access to antigen-bearing antigen-presenting cells. J. Exp. Med. 192:1105-1113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lawson, N. D., E. A. Stillman, M. A. Whitt, and J. K. Rose. 1995. Recombinant vesicular stomatitis viruses from DNA. Proc. Natl. Acad. Sci. USA 92:4477-4481. (Erratum, 92:9009, 1995.) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Leachman, S. A., M. Shylankevich, M. D. Slade, D. Levine, R. K. Sundaram, W. Xiao, M. Bryan, D. Zelterman, R. E. Tiegelaar, and J. L. Brandsma. 2002. Ubiquitin-fused and/or multiple early genes from cottontail rabbit papillomavirus as DNA vaccines. J. Virol. 76:7616-7624. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.McMurray, H. R., D. Nguyen, T. F. Westbrook, and D. J. McAnce. 2001. Biology of human papillomaviruses. Int. J. Exp. Pathol. 82:15-33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Puscas, L. 2005. The role of human papilloma virus infection in the etiology of oropharyngeal carcinoma. Curr. Opin. Otolaryngol. Head Neck Surg. 13:212-216. [DOI] [PubMed] [Google Scholar]
- 24.Ramsburg, E., N. F. Rose, P. A. Marx, M. Mefford, D. F. Nixon, W. J. Moretto, D. Montefiori, P. Earl, B. Moss, and J. K. Rose. 2004. Highly effective control of an AIDS virus challenge in macaques by using vesicular stomatitis virus and modified vaccinia virus Ankara vaccine vectors in a single-boost protocol. J. Virol. 78:3930-3940. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Reuter, J. D., B. E. Vivas-Gonzalez, D. Gomez, J. H. Wilson, J. L. Brandsma, H. L. Greenstone, J. K. Rose, and A. Roberts. 2002. Intranasal vaccination with a recombinant vesicular stomatitis virus expressing cottontail rabbit papillomavirus L1 protein provides complete protection against papillomavirus-induced disease. J. Virol. 76:8900-8909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Roberts, A., L. Buonocore, R. Price, J. Forman, and J. K. Rose. 1999. Attenuated vesicular stomatitis viruses as vaccine vectors. J. Virol. 73:3723-3732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Roberts, A., E. Kretzschmar, A. S. Perkins, J. Forman, R. Price, L. Buonocore, Y. Kawaoka, and J. K. Rose. 1998. Vaccination with a recombinant vesicular stomatitis virus expressing an influenza virus hemagglutinin provides complete protection from influenza virus challenge. J. Virol. 72:4704-4711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Roberts, A., J. D. Reuter, J. H. Wilson, S. Baldwin, and J. K. Rose. 2004. Complete protection from papillomavirus challenge after a single vaccination with a vesicular stomatitis virus vector expressing high levels of L1 protein. J. Virol. 78:3196-3199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Roden, R. B. S., M. Ling, and T. C. Wu. 2004. Vaccination to prevent and treat cervical cancer. Hum. Pathol. 35:971-982. [DOI] [PubMed] [Google Scholar]
- 30.Rose, J. K., and M. A. Whitt. 2001. Rhabdoviridae: the viruses and their replication, p. 1221-1240. In B. N. Fields and D. M. Knipe (ed.), Field's virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, PA.
- 31.Rose, N. F., P. A. Marx, A. Luckay, D. F. Nixon, W. J. Moretto, S. M. Donahoe, D. Montefiori, A. Roberts, L. Buonocore, and J. K. Rose. 2001. An effective AIDS vaccine based on live attenuated vesicular stomatitis virus recombinants. Cell 106:539-549. [DOI] [PubMed] [Google Scholar]
- 32.Rous, P., J. G. Kidd, and J. W. Beard. 1936. Observations on the relation of the virus causing rabbit papillomas to the cancers deriving therefrom. I. The influence of the host species and of the pathogenic activity and concentration of the virus. J. Exp. Med. 64:385-400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Saul, A., G. Lawrence, A. Smillie, C. M. Rzepczyk, C. Reed, D. Taylor, K. Anderson, A. Stowers, R. Kemp, A. Allworth, R. F. Anders, G. V. Brown, D. Pye, P. Schoofs, D. O. Irving, S. L. Dyer, G. C. Woodrow, W. R. Briggs, R. Reber, and D. Sturchler. 1999. Human phase I vaccine trials of 3 recombinant asexual stage malaria antigens with Montanide ISA720 adjuvant. Vaccine 17:3145-3159. [DOI] [PubMed] [Google Scholar]
- 34.Schlereth, B., J. K. Rose, L. Buonocore, V. ter Meulen, and S. Niewiesk. 2000. Successful vaccine-induced seroconversion by single-dose immunization in the presence of measles virus-specific maternal antibodies. J. Virol. 74:4652-4657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Seiler, P., M. A. Brundler, C. Zimmermann, D. Weibel, M. Bruns, H. Hengartner, and R. M. Zinkernagel. 1998. Induction of protective cytotoxic T cell responses in the presence of high titers of virus-neutralizing antibodies: implications for passive and active immunization. J. Exp. Med. 187:649-654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Selvakumar, R., L. A. Borenstein, Y. L. Lin, R. Ahmed, and F. O. Wettstein. 1995. Immunization with nonstructural proteins E1 and E2 of cottontail rabbit papillomavirus stimulates regression of virus-induced papillomas. J. Virol. 69:602-605. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Stanley, M. A. 2003. Progress in prophylactic and therapeutic vaccines for human papillomavirus infection. Expert Rev. Vaccines 2:381-389. [DOI] [PubMed] [Google Scholar]
- 38.Sundaram, P., and J. L. Brandsma. 1996. Rapid, efficient, large-scale purification of unfused, non-denatured E7 protein of cottontail rabbit papillomavirus. J. Virol. Methods 57:61-70. [DOI] [PubMed] [Google Scholar]
- 39.Sundaram, P., R. E. Tigelaar, and J. L. Brandsma. 1997. Intracutaneous vaccination of rabbits with the cottontail rabbit papillomavirus (CRPV) L1 gene protects against virus challenge. Vaccine 15:664-671. (Erratum, 16:655, 1998.) [DOI] [PubMed] [Google Scholar]
- 40.Sundaram, P., R. E. Tigelaar, W. Xiao, and J. L. Brandsma. 1998. Intracutaneous vaccination of rabbits with the E6 gene of cottontail rabbit papillomavirus provides partial protection against virus challenge. Vaccine 16:613-623. [DOI] [PubMed] [Google Scholar]
- 41.Tsang, K. Y., C. Palena, J. Yokokawa, P. M. Arlen, J. L. Gulley, G. P. Mazzara, L. Gritz, A. G. Yafal, S. Ogueta, P. Greenhalgh, K. Manson, D. Panicali, and J. Schlom. 2005. Analyses of recombinant vaccinia and fowlpox vaccine vectors expressing transgenes for two human tumor antigens and three human costimulatory molecules. Clin. Cancer Res. 11:1597-1607. [DOI] [PubMed] [Google Scholar]
- 42.Wettstein, F. O., and J. G. Stevens. 1980. Distribution and state of viral nucleic acid in tumors induced by Shope papilloma virus, p. 301-307. In M. Essex, G. Todaro, and H. zur Hausen (ed.), Viruses in naturally occurring cancers, vol. 7. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. [Google Scholar]