Pathogen-free C57BL/6 female mice, 6–8 weeks of age, were obtained from Charles River Laboratories (Wilmington, MA) and were housed under filter top conditions with water and food ad libitum. The B16-F10 melanoma cell line was obtained from American Type Culture Collection (ATCC, Manassas, VA). The melanoma cell line was cultured in Dulbecco's Modified Dulbecco's Medium (DMDM, Sigma, St. Louis, MO) supplemented with 10% heat-inactivated fetal calf serum (Hyclone), 2 mM L-glutamine (Gibco), 5 mM 2-mercaptoethanol (Gibco), penicillin and streptomycin (100 μg/ml;Gibco). Cells were incubated at 37°C/5% CO₂.

The tyrosinase-related protein peptides TRP2: 180–188 (SVYDFFVWL) and TRP2: 181–188 (VYDFFVWL), as well as the modified p53 peptide (p53: 232–240; KYICNSSCM) fused to PADRE (AKXVAAWTLKAAA, where X is cyclohexylalanine) were purchased from Dalton Chemical Laboratories Inc. (Toronto, ON, Canada). These peptides are presented by murine MHC-class I H-2K. The TRP2 and p53 peptides were stored as a 1 mg/ml stock solution containing DMSO to maintain solubility. Further dilutions for vaccine production were made using PBS.

All formulations of the vaccines contained either PADRE (AKXVAAWTLKAAA-OH; 25 μg/dose) and CpG ODN 1826 (5'-TCCAT

GACGTT

GACGTT

B16-F10 cells (10⁴ cells/mouse) were implanted subcutaneously in the left flank of C57BL/6 mice which were 6–8 weeks of age at time of challenge. Tumor were measured with calipers every 2–5 days. Data was reported as a percentage of tumor-bearing mice.

Four to six days after implantation of melanoma cells, mice (5 mice/group) received a single subcutaneous (s.c.) injection of VM containing TRP2: 180–188, TRP2: 181–188, or modified p53: 232–240 (25 μg of peptide/mouse). In addition, mice were injected with vaccine containing mixtures of two peptides (25 μg of each peptide/mouse). The mixtures contained either TRP2:180–188 and modified p53 or TRP2 181–188 and modified p53. VM formulations contained PADRE (25 μg/mouse) and CpG (50 μg/mouse) as adjuvants. The peptides and adjuvants were encapsulated in VM as previously described 1. In brief, lecithin and cholesterol in a ratio of 10:1 (0.2 g lecithin and 0.02 g cholesterol/dose) were dissolved in chloroform/methanol (1:1;v/v) and the solution was filter-sterilized using a PTFE 0.2 μm filter. Chloroform and methanol were removed under reduced pressure using a rotary evaporator and traces of the solvents were removed from the resulting thin lipid layer in vacuo. For liposome encapsulation, peptides with CpG and PADRE were dissolved in sterile PBS and the resulting solution added to the thin lipid layer with mixing to form liposomes. The resulting suspension of liposomes was emulsified in Montanide ISA51 (Seppic, France) by adding the liposome/PBS suspension to ISA51 to form a water-in-oil emulsion (PBS:ISA51; 1:1, v/v; 100 μl/dose). Control mice were injected with either vaccines that contained all components of test vaccines except liposomes or phosphate buffered saline (PBS, 100 μL/injection).

Activated antigen-specific splenocytes harvested from immunized C57BL/6 mice were detected using the BD ELISPOT kit following the manufacturer's instructions (BD Bioscience, San Diego, CA). Briefly, on day 8 post-immunization, a 96-well nitrocellulose plate was coated overnight at 4°C with capture antibody (anti-mouse IFN-γ) and then blocked with complete media. Splenocytes were added to wells at an initial concentration of 1 million cells/well in a volume of 100 μl. Cells in a dilution series were either non-stimulated or stimulated with relevant or irrelevant peptides (10 μg/ml).

PMA (5 ng/ml, Sigma) and ionomycin (500 ng/ml, Sigma), served as positive controls and irrelevant peptide and media alone served as negative controls. The plate was incubated overnight at 37°C/5% CO₂. The plate was then incubated with detection antibody (a biotinylated anti-mouse IFN-γ antibody), for 2 hours at room temperature. Unbound detection antibody was removed by washing and the enzyme conjugate (Streptavidin-HRP) was added. Following 1 hour incubation at room temperature, unbound enzyme conjugate was removed by washing and the plate was stained with an AEC substrate solution for 20 minutes. The plate was washed, allowed to air dry overnight, and foci of staining were counted using a magnifying lens.

The CD4⁺ T-cell helper response to PADRE was measured by ELISPOT. Eight days post-immunization, splenocytes were recovered from immunized mice and stimulated with PADRE (10 μg/ml). Production of IFN-γ was measured as described above.

Data are expressed as the mean ± SD and statistical significance was determined using paired two-tailed Student t-test (p < 0.05).

The B16-F10 melanoma model was used to demonstrate the therapeutic value of VM formulated with tumor-associated self antigens. Based on our observations that eradication of HPV TC1/A2 tumors was more effective when more than one CTL epitope-containing peptide was incorporated in VM 2, a mixture of 2 peptides formulated in a vaccine enhancement platform was used to eradicate B16-F10 melanoma tumors. All mice immunized against a mixture of TRP2:181–188 and modified p53: 232–240 were tumor-free by 21 days post-immunization (Fig. 1). Mice remained tumor-free until day 30–35, after which tumors regenerated in a number of mice. In this representative experiment, vaccinated mice remained tumor-free until day 33. Thereafter, tumors regenerated in two mice by day 35, in a third mouse by day 42 and in a fourth mouse by day 50 post-immunization. In two similar trials, immunization against a mixture of TRP2:180–188 and modified p53:232–240 in VM resulted in tumor rejection in 60–80% of treated mice within 26–32 days post-immunization, (data not shown) and these mice remained tumor-free at the end of the trial (day 40 post-immunization). In contrast, when TRP2 and p53 epitopes were combined with CpG and ISA51 (control vaccination, no liposomes), no more than 20% of mice remained tumor-free mice until day 25–30 and tumors regenerated in all vaccinated mice (data not shown).

Immunization against TRP2:180–188 alone plus CpG in VM eradicated tumors in 40% of the mice by 16 days post-immunization. These mice remained tumor-free for the entire monitoring period. Immunization against TRP2:181–188 alone also suppressed tumors in 40% of mice by 16 days post-immunization, but tumors regenerated in these mice by day 30 post-immunization (data not shown).

Immunization of mice against modified p53:232–240 alone delayed tumor development in one mouse but by day 21 post-immunization all mice were bearing tumors. Twenty percent of mice became tumor-free in the interval between day 25 and 27 post-immunization. All control mice injected with PBS developed tumors by day 14 post-immunization.

Previous reports have suggested that TRP2, and in particular TRP2 epitope 180–188, are poorly immunogenic 32425262728293031323334. In these studies, immunization against TRP2:180–188 required multiple administrations with CpG ODN to induce CTL responses. Even when complemented by PADRE and in IFA, this composition did not result in protection against B16-F10 tumor growth 2528. To assess the ability of VM to enhance the immune response to TRP2:180–188, mice (3–5/group) were immunized against these peptides or an irrelevant peptide in VM. Splenocytes from individual mice were harvested 8 days post-immunization and stimulated in vitro for 12 hours with TRP2:180–188. The number of TRP2:180–188-activated IFN-γ-producing T cells was determined by ELISPOT assay. A significant increase in the number of TRP2:180–188-specific IFN-γ-producing cells occurred by 8 days post-immunization (Fig. 2). This experiment indicated that VM without adjuvant could induce a CTL response against TRP2:180–188. The presence of CpG ODN in VM increased the CTL response against the epitope by approximately two-fold. TRP2:180–188 alone produced a background CTL response as did immunization against an irrelevant peptide. Similar results were obtained using splenocytes from mice vaccinated with TRP2:181–188 instead of TRP2:180–188 (data not shown). The CTL response to TRP2:181–188 with CpG ODN but without VM was approximately 20% of the CTL response to TRP2:181–188 with CpG ODN in VM. These results indicate that VM stimulates a robust CTL response to TRP2:180–188 and TRP2:181–188.

To assess the ability of VM to enhance the immune response to modified p53:232–240, mice (3–5/group) were immunized against modified p53:232–240 in VM, in VM without CpG, without VM liposomes, or against an irrelevant peptide in VM. Splenocytes from individual mice were harvested 8 days post-immunization and stimulated in vitro with modified p53:232–240. A significant increase in the number of modified p53:232–240-specific IFN-γ-producing cells was observed in mice immunized against modified p53:232–240 in VM by 8 days post-immunization (Fig. 3). This vaccine formulation without CpG produced background numbers of modified p53:232–240-specific splenocytes as did formulations that lacked the liposome component of VM (modified p53:232–240-PADRE-CpG and modified p53:232–240-PADRE). Immunization against an irrelevant peptide also produced similar background numbers of modified p53:232–240-specific splenocytes.

Melanoma cells express more than one tumor-associated protein simultaneously. TRP2 is a tumor-associated protein limited to melanoma, whereas, p53 is over-expressed in a wide variety of cancers. For example, the p53 gene is commonly mutated in lung, colon, and breast cancers. The observation that immunization against a mixture of peptides derived from TRP2 and p53 was more effective in eradicating B16-F10 melanoma tumors than immunization against each of these peptides alone suggests that use of VM induces a CTL response against more than one CTL epitope simultaneously. To evaluate the ability of VM to raise a simultaneous immune response against two targets, mice (3/group) were immunized once against a mixture of TRP2:180–188 and modified p53:232–240 peptides, with and without VM, or with and without CpG ODN (Fig. 4). All vaccine formulations contained water-in-oil emulsions. Formulations without VM contained all vaccine ingredients except liposomes. Spleens were collected 8 days post-immunization and the number of IFN-γ producing cells was measured by ELISPOT assay. A single immunization with VM containing both TRP2:180–188 and modified p53:232–240 antigens stimulated a robust IFN-γ response against both antigens simultaneously. Spleens from mice immunized with VM contained similar numbers of TRP2:180–188-specific and modified p53:232–240-specific IFN-γ producing cells. In contrast, spleens from mice immunized by a single vaccination of a mixture of TRP2:180–188 and modified p53:232–240 without VM produced a robust TRP2:180–188-specific response but a weak modified p53:232–240-specific response. Vaccination against a mixture of TRP2:180–188 and modified p53:232–240 without CpG resulted in a poor TRP2:180–188-specific IFN-γ response and a slightly better modified p53:232–240-specific response. Without VM, both TRP2:180–188- and modified p53:232–240-specific IFN-γ responses were poor.

CD4⁺ T-cell help is essential for the differentiation and expansion of CTLs 35, as well as their maturation into functional memory CTLs 36. To achieve CD4⁺ T-cell help, a universal T helper epitope, PADRE was used in VM 37. A strong PADRE-specific CD4+ T cell help was induced in mice immunized against TRP2:180–188 with and without VM (Fig. 5). These results along with figures 1 and 2 show the simultaneous induction of immune responses to multiple MHC class I and II peptides when encapsulated in VM.