Female 6–8-week-old C57BL/6 mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Animals were handled under aseptic conditions in microisolator cages within the Central Animal Facility at the University of Pittsburgh per an Institutional Animal Care and Use Committee-approved protocol, and in accordance with recommendations for the proper care and use of laboratory animals.

Glioma cell lines KR129, KR130, KR233 and KR158D were derived from spontaneously arising gliomas in F1 between B6 × CBA (KR129), B6 × SJL (KR130), and C57BL/6 (KR233 and KR158D)-background NPcis mice that express mutations in two tumor-suppressor genes, Nf1 and Trp53 12. B16 melanoma, NPcis-derived glioma cells, GL261 glioma and MCA205 sarcoma (H-2^b) cell lines were cultured in complete medium (CM) [RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum, 100 units/ml penicillin, 100 μg/ml streptomycin, and 10 mM L-glutamine (all reagents from Life Technologies, Inc., Grand Island, NY)] in a humidified incubator in 5% CO₂ at 37°C.

The procedure used to generate DCs has been previously described 13. Briefly, C57BL/6 bone marrow cells were cultured in CM supplemented with 1000 units/ml recombinant mouse granulocyte/macrophage colony-stimulating factor and recombinant mouse interleukin-4 (Schering-Plough, Kenilworth, NJ) at 37°C in a humidified, 5% CO₂ incubator for 7 days. DCs were then isolated at the interface of 14.5% (w/v) metrizamide (Sigma, St. Louis, MO) in CM discontinuous gradients by centrifugation. DCs typically represented >90% of the harvested population of cells based on morphology and expression of the CD11b, CD11c, CD40, CD54, CD80, CD86, and class I and class II MHC antigens as assessed using flow cytometry (data not shown).

The protein sequences of mEphA2 was obtained from GenBank and analyzed for H-2K^b-, H-2D^b-, and I-A^b-binding binding motifs using

BIMAS

Day 7 DCs were pulsed with 10 μM each of the indicated peptides for 4 hours at 37°C, as previously described 13. Cells were then washed twice with Hank's balanced salt solution (HBSS), with animals receiving injections of the indicated numbers of peptide-pulsed DCs in 0.1 ml HBSS s.c.

In the s.c. model, animals were injected with the indicated numbers of tumor cells in the right flank. Anti-tumor responses were assessed based on comparative longitudinal measurements of tumor area. In the lung metastasis model, animals were injected i.v. with 2 × 10⁵ B16-BL6 tumor cells on day 0, and they subsequently received s.c. vaccinations of peptide-loaded DCs on days 3, 10 and 17. Animals were sacrificed on day 28 post-tumor injection and analyzed for the assessment of pulmonary metastases by enumerating the number of surface tumor-nodules.

Protein lysates isolated from normal mouse brain, spleen, liver, lung, heart, skeletal muscle, and mouse tumor lines were separated by SDS-PAGE, blotted onto nitrocellulose and analyzed for expression of mEphA2 using EphA2 monoclonal antibody (C-20 Ab; Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Blots were imaged on Kodak X-Omat Blue XB-1 film (NEN Life Science Products, Boston, MA) after using horseradish peroxidase (HRP)-conjugated goat anti-rabbit Ig (Biorad, Hercules, CA) and the Western Lighting™ chemiluminescence detection kit (Perkin Elmer, Boston, MA).

Single cell suspensions of SPCs were cultured at 2 × 10⁶ cells/ml with 2 μg/ml mEphA2_671–679 or mEphA2_682–689 in presence of 10 U/ml human IL-2 (Chiron, Emeryville, CA), 50 μM 2-mercaptoethanol (Sigma), and 50 μM N^G mono-methyl-L-arginine (Cyclopss, Salt Lake City, UT) in 24 wells plates (Corning, Corning, NY) for 5 days. Specific CTL activity was determined in 4 h ⁵¹Cr release assays against the indicated target cells, as previously described 16.

Eight-week-old mice were injected s.c. twice (days -14 and -7) with DCs loaded with either EphA2-derived peptides (EphA2 _{671–679/30–44}) or OVA-derived peptides (OVA _{257–264/265–280}). Each animal then received 400 μl of Matrigel (BD Biosciences) supplemented with 400 ng/ml vascular endothelial growth factor (VEGF) in the dorsal area 17. The animals were sacrificed 10 days later, then the plugs were removed and photographed.

Comparative numbers of lung metastasis, growth of s.c. tumors and T cell responses were compared by Student's t test for two samples with unequal variances. Statistical significance was determined at p value <0.05.

We evaluated the expression of mEphA2 in H-2^b-syngeneic murine tumor cell lines and normal organs from C57BL/6 mice by Western Blotting (Figure 1). MCA205 sarcoma, the NPcis mice-derived spontaneously developed glioma cell lines KR129, KR130, KR233 and KR158D (Figure 1A), as well as, the GL261 glioma (Figure 1B) expressed detectable but variable levels of mEphA2 protein. In marked contrast, none of melanoma lines tested, including B16 melanoma and its more metastatic sub-clones B16V1 and B16BL6, expressed detectable levels of mEphA2 (Figure 1A). The control β-actin staining on the same blots demonstrated that equal amount of protein was loaded in each lane. With regard to the normal organs examined, moderate levels of mEphA2 expression were identified in the spleen, liver and lung, whereas no detectable expression was observed in brain, heart and skeletal muscle (Figure 1B).

We have previously characterized the immunogenicity of the H-2D^b-binding mEphA2_671–679, H-2K^b-binding mEphA2_682–689, and I-A^b-binding mEphA2_30–44 epitopes based on delayed-type hypersensitivity responses in B6 mice (Storkus et al., unpublished results). We immunized syngeneic C57BL/6 mice twice s.c. with syngenic DCs loaded with the K^b-binding and the I-A^b-binding epitopes, or DCs loaded with the D^b-binding and the I-A^b-binding epitopes on a weekly basis. Control animals received DCs loaded with the cancer-irrelevant K^b-binding OVA_257–264 and the I-A^b-binding OVA_265–280 epitopes 18. One week after the secondary immunization, the animals were sacrificed, and SPCs were harvested. The SPCs were then stimulated in vitro with the MHC class I peptide that was used for in vivo immunization, in the presence of 10 U/ml hIL-2, for 7 days. SPCs from control animals immunized with OVA peptides were stimulated in vitro with the mEphA2_671–79 to provide an index for a control, "primary" level of specific CTL induction. CTL activity against mEphA2-expressing, and H-2^b-syngeneic MCA205 sarcoma, GL261 glioma and KR158D glioma cells were assessed by standard ⁵¹Cr-release assays (Figure 2). All three mEphA2+ cell lines demonstrated susceptibility to CTLs generated by the immunizations with K^b- or D^b-binding mEphA2 derived epitopes. With 20 h – incubation time, the percent lysis by EphA2-induced CTLs increased remarkably in comparison to 4 h-incubation regimen, whereas the lysis by OVA-induced CTLs do not, suggesting that the specific lysis can be more appreciated with the prolonged incubation time (up to 20 h) for the CTLs raised against EphA2-epitopes. B16 melanoma and YAC cells, which were used as negative control target cells, displayed constantly less than 10% of specific lyses in all groups (data not shown). These data support the conclusion that these mEphA2-derived peptides may serve as effective in vivo immunogens for anti-tumor T cell activation.

We evaluated the efficacy of protective and therapeutic immunizations with mEphA2-derived peptides in s.c. tumor models. These included the mEphA2-positive MCA205 sarcoma and the mEphA2-negative B16 melanoma models. First, animals bearing established s.c. MCA205 tumors received therapeutic vaccinations with mEphA2-derived CTL (H-2K^b or H-D^b presented) + Th epitope peptides on days 3 and 10 following tumor inoculation (5 animals/group). Control animals received DCs loaded with the OVA-derived CTL + Th epitopes. As displayed in Figure 3A, vaccination with either combination of mEphA2-derived CTL (H-2K^b or H-D^b presented) + Th epitope peptides significantly inhibited the growth of tumors in comparison to control vaccinations with irrelevant (OVA)-derived T cell epitopes.

We next examined whether vaccination with mEphA2-derived peptides could induce anti-tumor effects against mEphA2-negative tumors, expecting that this would not be the case. Since s.c. B16 melanomas grow aggressively, we employed a pre-immunization model to assess the anti-tumor response. Syngeneic C57BL/6 mice were pre-immunized with either mEphA2_671–679, mEphA2_682–679, or mEphA2_30–44 peptides on days -14 and -7 before s.c. injection with 5 × 10⁴ B16 cells on day 0 (5 animals/group). The control group received irrelevant OVA_257–264 instead of mEphA2-derived peptides. Tumor size and the number of animals that had measurable tumors are assessed on day 21 (Figure 3B). All control animals receiving OVA-immunization developed large tumors, whereas, surprisingly, 3 of 5 animals in each of mEphA2-derived peptide-treated groups rejected the tumor growth. Indeed, the average tumor size in each of mEphA2-immunized groups was significantly smaller than the control groups (P < 0.01). As B16 melanoma cells did not express mEphA2 protein in vitro (Figure 1A), these data suggest that vaccination with mEphA2-derived peptides may induce in vivo CTL activity not only directly against mEphA2 expressed in tumors, but also against mEphA2 expressed by other critical components of tumor-structure, such as tumor vascular endothelial cells. With regard to EphA2 expression in vivo, we attempted immunohistochemistry on s.c. B16 tumors to directly validate EphA2 expression. Although deposition of endogenous melanin in B16 tissues posed interference for the substrate-development, we did not observe EphA2-specific staining in s.c. B16 tumor tissues (data not shown).

We next evaluated a lung metastasis model using B16-BL6 melanoma cell line. Syngeneic C57BL/6 mice received 2 × 10⁵ B16-BL6 melanoma cells via tail vein injection. The animals subsequently received s.c. immunizations with 1 × 10⁶ DCs loaded with mEphA2 CTL (H-2K^b or H-D^b presented) + Th on days 3, 10 and 17 (5 animals/group). Control animals received DCs loaded with the OVA CTL + Th epitopes. On day 28, the animals were sacrificed, and the lungs harvested from each animal. The number of surface pulmonary metastases and lung weight were measured and compared between the groups. As demonstrated in Figure 4A and 4B, both of mEphA2-immunization regimens resulted in remarkable suppression of B16-BL6 lung metastases in comparison to the OVA-immunized control group in this early-stage treatment model.

VEGF has been implicated to be a major mediator of tumor-angiogenesis 1920, and EphA2-blockade has been shown to inhibit VEGF-induced angiogenesis 21. To examine whether immunizations with mEphA2-derived peptides inhibit VEGF-induced angiogenesis, Matrigel plugs containing VEGF were implanted in pre-immunized mice and recovered 10 days later. Macroscopic analysis revealed that the plugs from mice pre-immunized with mEphA2-derived peptides were much paler than those from control mice (Figure 4C). These results suggest that vaccinations with mEphA2-derived T cell epitopes may induce potent anti-tumor responses in mEphA2-deficient tumor models through inhibition of tumor angiogenesis.