Surrogate Tumor Antigen Vaccination Induces ... - Cancer Research

Research Article

Surrogate Tumor Antigen Vaccination Induces Tumor-Specific Immunity and the Rejection of Spontaneous Metastases Jennifer D. Lewis, Michael H. Shearer, Ronald C. Kennedy, and Robert K. Bright Department of Microbiology and Immunology and the Southwest Cancer Treatment and Research Center, Texas Tech University Health Sciences Center, Lubbock, Texas

Abstract The nonimmunogenic 4T1 murine mammary carcinoma model and a model surrogate tumor antigen (sTA) were employed to explore the possibility of inducing tumor-specific immunity through active immunization in the absence of defined tumor-associated antigens. Immunization of naive mice with protein-based sTA resulted in protection from s.c. challenge, with 4T1 modified to express the sTA (4T1.sTA), or from a sTA-expressing unrelated tumor cell line (mKSA). Immunization had no effect on parental 4T1 tumor growth or the formation of parental 4T1 spontaneous lung metastases. Mice that were sTA immunized and successfully rejected 4T1.sTA challenge also rejected a subsequent challenge in the contralateral flank with parental 4T1 and strikingly prevented the formation of spontaneous parental 4T1 lung metastases. The rejection of parental 4T1 seemed to be specific for and associated with unknown 4T1 tumor-associated antigens, because rejection of mKSA did not induce cross-protection against a challenge with parental 4T1. To evaluate the effect of this vaccine approach on established disease, mice were simultaneously challenged on day 0 with 4T1.sTA and parental 4T1 in contralateral flanks and then immunized on days 3, 10, 17, and 24 with sTA protein. Tumor growth and metastasis were delayed in four of five animals, and 20% (2 of 5) of the animals were tumor free at the completion of the experiment. Together, these data suggest that prior vaccination with a sTA followed by inoculation with poorly immunogenic tumor cells modified to express the sTA activates determinant spreading and the induction of systemic tumor immunity resulting in indigenous tumor rejection. (Cancer Res 2005; 65(7): 2938-46)

Introduction The field of cancer vaccine development is growing rapidly and has become the focus of much intense investigation. Cancer vaccines have evolved from crude tumor cell lysates given with an adjuvant to the sophisticated application of ex vivo engineered dendritic cells. Indeed, proteins, peptides, DNA, viral vectors, heat shock proteins, and dendritic cells as well as various combinations and permutations of these modalities have been and are being explored and refined as vaccines against cancer (1, 2). In the midst of this intense study, one thing is becoming increasingly clear: the most effective vaccines may be those composed of a defined tumor-associated antigen(s) (TAA). Thus, the need to discover,

Requests for reprints: Robert K. Bright, Department of Microbiology and Immunology, Texas Tech University Health Sciences Center, 3601 4th Street, MS 6591, Lubbock, TX 79430. Phone: 806-743-4592; Fax: 806-743-2334; E-mail: [email protected]. I2005 American Association for Cancer Research.

Cancer Res 2005; 65: (7). April 1, 2005

define, and validate additional TAA for cancers to include prostate, breast, and lung cancers remains an obstacle for effective vaccine development (3). To this end, new vaccine methods that circumvent the need for in hand, defined TAA could be extremely valuable. The BALB/c-derived 4T1 mammary carcinoma model is a unique nontransgenic model of metastatic breast cancer based on explantable, syngeneic, clonal tumor cells (4). The growth characteristics, lack of immunogenicity, and aggressive, spontaneous metastatic properties of this model closely resemble those characteristics observed for human breast cancer (5). 4T1 tumor cells are relatively easy to culture and form tumors in healthy, immunocompetent BALB/c mice following s.c. inoculation with as few as 5,000 to 7,000 cells (6, 7), making this a useful in vivo model to study the effects of active immunization against aggressive, nonimmunogenic cancers for which no TAAs have been defined. Epitope or determinant spreading as it relates to specific immune responses against a given antigen was first described as the dispersion of immunologic specificities from a single dominant epitope to subdominant epitopes (8–10) within the original antigenic target (intramolecular determinant spreading) or to epitopes carried by distinct antigens (intermolecular determinant spreading; ref. 11). Early studies showed that clones of specific T cells were responsible for driving determinant spreading in models of autoimmunity (12). Driver T-cell clones as the mechanism of determinant spreading initiation in autoimmunity were later supported by murine models of tumor vaccination. A vaccine composed of a single peptide epitope from chicken ovalbumin delivered on irradiated RMAS cells initiated a CTL response that resulted in the rejection of ovalbumin-expressing tumors and the generation of CTL with specificities for additional ovalbumin epitopes (intramolecular determinant spreading) as well as the generation of CTL with specificities for epitopes on distinct antigens (intermolecular determinant spreading; ref. 13). In a separate study, a single peptide, P1A, induced CTL that rejected P1A+ tumor challenge and subsequently yielded CTL activity and tumor protection against challenge with a P1A variant, which resulted from determinant spreading to a second defined epitope (P1E; ref. 14). These studies showed that peptide vaccines directed at the induction of CTL responses are capable of inducing tumor immunity and that determinant spreading occurs, resulting in the expansion of the immune response to additional epitopes in murine tumor models of tumor protection. In the present study, we employed the aggressive, tumorigenic, and nonimmunogenic 4T1 murine mammary carcinoma model and SV40 large tumor antigen (Tag) as a model surrogate tumor antigen (sTA) to explore the possibility of inducing tumor-specific immunity through active immunization in the absence of a defined TAA and to address whether vaccination with a soluble whole protein would activate determinant spreading and tumor rejection in vivo. Mice immunized with sTA rejected challenge with 4T1

2938

www.aacrjournals.org

Surrogate Antigen-Based Tumor Vaccines

tumor cells modified to express the sTA, which led to subsequent rejection of parental 4T1 tumor cells. Rejection of parental 4T1 tumor formation seemed to be specific for and associated with unknown 4T1 TAA, because rejection of sTA-expressing unrelated mKSA kidney-derived tumor cells did not induce cross-protection against a subsequent contralateral challenge with parental 4T1 tumor cells. Moreover, in a tumor treatment experiment where mice were simultaneously inoculated in contralateral flanks with parental 4T1 tumor cells and sTA-expressing 4T1 tumor cells followed by sTA vaccination, 20% of the animals were tumor free at the completion of the experiment. Taken together, these data suggest that immunization with a whole, soluble protein-based sTA induces intermolecular determinant spreading, resulting in the induction of systemic tumor immunity and the subsequent rejection of s.c. tumor growth and prevention of the formation of lethal, spontaneous metastases.

Materials and Methods Surrogate tumor antigen. Recombinant SV40 Tag served as a model sTA and was generated in the Sf9 insect cell line using the baculovirus AcNPV expression system. Recombinant Tag was extracted from insect cells and immunoaffinity purified by methods described previously (15). Purity was assessed by SDS-PAGE and silver staining (16) and the concentration was determined using an absorption coefficient at 280 nm for a 1% solution. Mice and cell lines. Female, 6- to 8-week-old female BALB/c mice were purchased from the NIH (Frederick, MD). All animals were cared for and treated according to Institutional Animal Use and Care Committee guidelines at Texas Tech University Health Sciences Center (Lubbock, TX). The tumorigenic murine mammary carcinoma cell line 4T1 was a kind gift from Fred Miller (Karmanos Cancer Institute, Detroit, MI). The tumorigenic SV40-transformed BALB/c murine kidney cell line designated mKSA (17) was used as a control for tumor growth and immunity following sTA immunization (18). Cells were cultured in RPMI 1640 (Fisher Scientific, Pittsburgh, PA) supplemented with 10% heat-inactivated fetal bovine serum, 2 mmol/L L-glutamine, 250 ng/mL fungizone, 50 IU/mL penicillin, 50 Ag/mL streptomycin, 50 Ag/mL gentamicin sulfate, and 10 mmol/L HEPES. The 4T1.sTA cell line was generated by transfecting parental 4T1 tumor cells with a plasmid encoding SV40 Tag as a model sTA using LipofectAMINE Plus reagent (Invitrogen Corp., Carlsbad, CA) and methods described previously (19). RNA isolation and reverse transcription-PCR. Briefly, total RNA was isolated from mKSA (sTA positive), 4T1.sTA (sTA transfected), P815 (sTA negative), and 4T1 (sTA negative) using Trizol reagent (BioWhittaker, Walkersville, MD) and 1 Ag was reverse transcribed to obtain cDNA. The primers and PCR conditions were described previously (19). PCR results were visualized on a 2% agarose gel containing ethidium bromide; assessment of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression served as an internal reference control. Western analysis. The cell lines mKSA, P815, 4T1.sTA, and 4T1 were grown to confluency in RPMI supplemented as described above. Cells were harvested and counted and whole cell protein lysates were prepared for 4T1, 4T1.sTA, P815, and mKSA at a concentration of 1 108 cell equivalents/mL using methods described previously (20). Briefly, samples were loaded on a standard reducing SDS-PAGE (4% stacking, 7% resolving) and run at 200 V for 1 hour. Next, the gel-separated samples were electrophoretically transferred to nitrocellulose and the membrane was blocked in PBS containing 1% normal goat serum, 1% milk, and 0.02% Tween for 2 hours at room temperature. To visual the expression of the model sTA (SV40 Tag), the membrane was probed with the anti-SV40 Tag monoclonal antibody (mAb) Pab 405 overnight at 4jC (19) and then washed thrice with PBS0.02% Tween, following which goat anti-mouse IgG-specific horseradish peroxidase reagent was added at a 1:2,000 dilution for 2 hours at room temperature. Finally, the membrane was washed and developed with 3,3Vdiaminobenzidine reagent as a substrate (Sigma, St. Louis, MO).


Intracellular detection of surrogate tumor antigen in tumor cell lines. Intracellular detection of sTA in viable cell lines was done using a commercially available Cytofix/Cytoperm kit (BD Biosciences, San Diego, CA) with the following modifications. Briefly, actively growing tumor cell lines were harvested, adjusted to 500,000 cells/mL, washed twice with 1 mL ice-cold PBS, pelleted, and then resuspended in 875 AL ice-cold PBS. Next, ice-cold fixation solution (125 AL) was added to the cells followed by 1-hour incubation at 4jC. Following incubation, cells were pelleted, the supernatant was aspirated, and 1 permeabilization/wash solution (1 mL) was added for 20 minutes at 37jC in the dark. Permeabilization/wash buffer (1) was used in all subsequent washes and dilutions to maintain permeabilization of cells. Following permeabilization, cells were washed once and pelleted, the supernatant was aspirated, and FITC-conjugated SV40 Tag–specific mAb (Pab 405, IgG1; 100 AL; 1 Ag final concentration) or irrelevant FITC-conjugated antibody (IgG1) was added with incubation for 30 minutes at 4jC in the dark. Finally, cells were washed, resuspended in 1 permeabilization/wash buffer (1 mL), and stored in the dark. All assays were analyzed by flow cytometry (FACScan, Becton Dickinson, Franklin Lakes, NJ) within 2 hours of staining. Immunization and tumor protection. Mice were immunized four times i.p. with sTA in alum at 1-week intervals at a concentration of 5 Ag/ animal for the first three injections and a concentration of 15 Ag/animal for the final injection. Control groups of mice were immunized with either PBS or alum alone using identical methods and schedules (18). Mice in all groups were bled from the dorsal tail vein before immunization and 7 days following each immunization. Three tumor cell lines were used for s.c. tumor challenge at the following doses: mKSA (1 106), 4T1.Tag (1 105), and 4T1 (1 104). Following harvest and counting, tumor cells were resuspended in versene (PBS-EDTA, Fisher Scientific) to prevent aggregation and injected s.c. in 100 AL. Primary challenge was given in the right flank with either mKSA (sTA positive) or 4T1.sTA followed 2 weeks later by a secondary challenge in the left flank of all animals with parental 4T1 tumor cells. Tumor size was recorded by measuring perpendicular angles using calipers every 2 to 3 days. Tumor volume (mm3) was calculated using the equation: (a b 2) / 2, where b is the smaller of the two measurements. To evaluate metastasis, lungs were harvested and injected with an India ink solution and tumor nodules were enumerated as described below. Immunization and tumor treatment. Parental 4T1 and sTA-modified 4T1 (4T1.sTA) tumor cells at the following doses were used in the therapeutic immunization model: 4T1.sTA tumor cells (1 104) and 4T1 tumor cells (1 104). Tumor cells were harvested, counted, and processed as described above. Mice were challenged s.c. simultaneously in contralateral flanks on day 0. On days 3, 10, 17, and 24, mice were immunized i.p. with sTA in alum using doses of 5, 5, 10, and 15 Ag, respectively. Immunization with alum alone or PBS served as controls. Tumor size was recorded by measuring perpendicular angles with calipers every 2 to 3 days. Tumor volume (mm3) was calculated using the equation: (a b 2) / 2, where b is the smaller of the two measurements. To evaluate metastasis, lungs were harvested and injected with an India ink solution and tumor nodules were enumerated as described below. Enumeration of 4T1 spontaneous lung metastases. To analyze tumor metastasis to the lungs, mice were sacrificed and their lungs were removed and injected with India ink to visualize individual tumor nodules. Briefly, an India ink solution was injected through the trachea and allowed to fill the lungs. The lungs were removed and placed in Fekete’s solution for destaining. Tumor nodules do not absorb India ink, which results in the normal lung tissue staining black and the tumor nodules remaining white. Tumor nodules were counted blindly and size was noted by three independent investigators. Immunoassay for detection of surrogate tumor antigen–specific antibodies. An ELISA was employed to detect the presence of sTA (Tag)– specific antibodies in the sera of mice following immunization using methods described previously (21, 22). Briefly, 96-well microtiter plates (Nunc Maxisorp, Fisher Scientific) were coated overnight at 4jC with 50 AL of the model sTA, recombinant SV40 Tag, in PBS at a final concentration of 200 ng/ well. Plates were blocked with 10% normal goat serum/PBS for 1 hour at 37jC to prevent nonspecific binding. Next, 2-fold dilutions (beginning dilution,

2939

Cancer Res 2005; 65: (7). April 1, 2005

Cancer Research 1:50; 50 AL) of mouse sera were added to individual wells in duplicate and plates were incubated overnight at 4jC. Antibody-antigen reactivity was detected using 50 AL goat anti-mouse IgG Fc-specific antibody conjugated to horseradish peroxidase (Sigma). Plates were washed between steps with 0.02% Tween-PBS. Reactions were developed with 100 AL peroxidase substrate [2,2V-azino-di(3-ethyl)benzthiazoline sulfonic acid] containing 0.01% hydrogen peroxide. Enzyme substrate reactions were terminated following the addition of 5% (w/v) SDS (100 AL). Absorbance was measured at 405 nm using the Victor3 plate reader (Wallac, Perkin-Elmer, Boston, MA). T-cell culture and assay for IFN-; production. T cells were isolated by culturing Ficoll-separated spleen cells with irradiated 4T1 tumor cells in the presence of interleukin (IL)-2 (10 ng/mL), IL-7 (5 ng/mL), and IL-12 (5 ng/mL; Peprotech, Rocky Hill, NJ) at 37jC for 5 to 7 days. Assessment of cytokine secretion by antigen-specific T-cell cultures (generated as described above) was accomplished by applying culture supernatants to commercially available sandwich ELISA kits for IFN-g detection (R&D Systems, Minneapolis, MN). Supernatants were harvested from 24-hour cultures of T cells (1 106 cells/mL in 200 AL medium in 96-well plates) in medium alone compared with T cells cultured with 4T1 tumor cells (1:1 ratio). To confirm MHC-I-restricted tumor recognition, CTL blocking assays were done by incubating 4T1 tumor cells with anti-H-2d (Kd, Ld, and Dd) or anti-H-2b (Kb; control) mAb before incubation with CTLs. Briefly, mAb (10 AL) in PBS ( final concentration, 30 Ag/mL) was added to individual wells of 96-well round-bottomed plates in triplicate. Next, 4T1 tumor targets (100 AL) were added to each well and incubated for 30 minutes at room temperature. Finally, effectors (100 AL) were added to the appropriate wells and the plates were incubated for 24 hours at 37jC. Assays were analyzed using the Victor3 plate reader. Graphical data in Fig. 3D represent the mean of triplicate values for CTL incubated with 4T1 tumor cells after subtraction of the values for CTLs alone.

Results Generation of surrogate tumor antigen–positive murine tumor cell lines. To generate a tumor cell line expressing the model sTA, parental 4T1 tumor cells were transfected with a plasmid encoding SV40 Tag using LipofectAMINE Plus reagent; the resulting cell line was designated 4T1.sTA. Expression of mRNA encoding the model sTA by 4T1.sTA was confirmed by reverse transcription-PCR (RT-PCR) using methods published previously by our laboratory (19). sTA expression was not detected in either the untransfected parental 4T1 cell line or the sTA-negative P815 plasmacytoma cell line (Fig. 1A). The SV40-transformed BALB/c, kidney-derived tumor cell line mKSA, served as a positive control for expression of the model sTA (SV40 Tag; Fig. 1A). Protein expression of the sTA was confirmed by Western blot analysis. A band of f90 kDa (23) was observed for the transfected 4T1.sTA tumor cell line as well as for the positive control mKSA tumor cell line (Fig. 1B). Examination of the parental 4T1 cell line before sTA transfection revealed no detectable levels of sTA protein expression by Western blot (Fig. 1B). To further evaluate the expression of sTA protein by 4T1.sTA cells, intracellular staining and flow cytometry were done using a FITC-labeled mAb (Pab 405) with specificity for the model sTA (Fig. 1C). Consistent with RTPCR and Western analysis, 4T1.sTA expressed the sTA at detectable levels (III) compared with lack of expression by the parental 4T1 tumor cell line (II), albeit much less than the positive control tumor cell line mKSA (I). The disparity in expression between 4T1.sTA and mKSA is likely because mKSA was generated by transformation with SV40 resulting in multiple copies of the gene for Tag being integrated into the cell genome. Tumor protection following immunization with surrogate tumor antigen. Next, we evaluated whether expression of the model sTA affected the growth kinetics of 4T1 tumor cells in vivo

Cancer Res 2005; 65: (7). April 1, 2005

and whether immunization with sTA protein in alum would result in rejection of 4T1.sTA tumor cells compared with parental 4T1 tumor cells. In brief, groups of five mice were immunized i.p. with 5 Ag purified recombinant sTA in alum in a total volume of 100 AL once every 7 days for 3 weeks followed by 15 Ag sTA in alum on the fourth week. Following immunization with sTA, mice were monitored for the induction of sTA-specific immunity by collecting serum samples and measuring anti-sTA antibody titers using methods and an ELISA generated in and reported by our laboratory (22). All three groups of mice (total of 15 mice) immunized with sTA in alum generated detectable IgG antibody titers specific for the sTA 2 weeks following the final immunization with nearidentical end point titers of 1:2,000. In contrast, mice immunized with PBS or alum alone showed no detectable IgG antibody titers to the sTA when sera were analyzed by ELISA with the sTA on the solid phase. Following immunization with sTA in alum, alum alone, or PBS, mice were challenged with either mKSA, parental 4T1, or 4T1.sTA at an inoculation dose of 5 105, 1 104, or 1 105, respectively, and tumor growth was measured using calipers as described in Materials and Methods. Mice challenged with mKSA tumor cells following immunization with sTA in alum did not develop tumors (Fig. 2A), whereas mice that were immunized with alum alone or PBS developed tumors at approximately the same rate (Fig. 2A-C). All the mice challenged with parental 4T1 tumor cells following immunization with either sTA in alum, alum alone, or PBS developed tumors at similar rates regardless of the immunization they were given (Fig. 2B). Importantly, mice challenged with 4T1.sTA tumor cells following immunization with sTA in alum did not develop tumors compared with mice immunized with alum alone or PBS, which developed tumors at similar rates (Fig. 2C). These data establish that immunization with a soluble, whole protein sTA induces immune responses capable of rejecting lethal challenge with sTA-positive tumor cells as well as show that transfection of parental 4T1 tumor cells with the sTA had no effect on in vivo tumor growth of the 4T1 cell in unimmunized mice. 4T1 tumor cells spontaneously metastasize to the lungs forming lethal tumor burdens. To evaluate metastasis, lungs were harvested and stained with an India ink solution and tumor nodules were enumerated. S.c. 4T1.sTA tumor challenge following immunization with sTA in alum not only resulted in rejection of the s.c. tumor but also prevented the development of spontaneous lung metastases (Table 1). In contrast, if mice were immunized with sTA in alum and challenged with parental 4T1 tumor cells, lung tumor nodules were present ranging from 10 to 18 large visible tumor nodules (Table 1; mice had to be sacrificed by day 42 following primary challenge with parental 4T1 tumor cells due to morbidity associated with the tumor burden). Therefore, immunization with sTA in alum induced sTA-specific immunity that was capable of preventing tumor formation and metastasis by nonimmunogenic 4T1 tumor cells modified to express the model sTA. Tumor protection as a result of surrogate tumor antigen– induced determinant spreading. To determine whether a sTAbased vaccine could be used to induce determinant spreading and rejection of indigenous tumors in the absence of known, defined TAA, groups of mice were immunized with sTA in alum and challenged with either mKSA tumor cells or 4T1.sTA tumor cells (as described for Fig. 2). Mice that were immunized with sTA in alum and rejected a primary s.c. challenge with the control sTA-positive, kidney-derived mKSA tumor cell line failed to reject a subsequent s.c. challenge with parental 4T1 tumor cells given 2 weeks later in

2940



Figure 1. Expression of the model sTA in murine tumor cells. A, RT-PCR demonstrating expression of the model sTA (SV40 Tag) in 4T1 murine tumor cells following transfection with a plasmid encoding the full-length cDNA for SV40 Tag. mKSA is a SV40-transformed, SV40 Tag–positive murine kidney-derived tumor of BALB/c origin and served as a positive control. P815 is a Tag-negative murine tumor cell line. GAPDH served as an internal reference control. B, Western blot analysis of murine tumor cell lysates generated from 1 108 cell equivalents demonstrating protein expression of sTA using a mAb (Pab 405) with specificity for the model sTA (SV40 Tag). The model sTA typically blots as two bands f90 kDa. mKSA is a sTA-positive murine kidney-derived tumor of BALB/c origin and served as a positive control. The sTA-negative tumor cell line P815 served as a negative control. GAPDH served as a control for protein loading. C, expression of sTA protein by flow cytometry. Intracellular sTA detection was accomplished using a Cytofix/Cytoperm reagent and a mAb (Pab 405) with specificity for SV40 Tag. Left, staining of mKSA tumor cells as a positive control; center, lack of staining of parental 4T1 tumor cells; right, staining of 4T1 tumor cells modified to express the model sTA (SV40 Tag). Solid, fluorescence of cells alone; solid line, staining with an isotype-matched control antibody (IgG); dashed line, staining with the FITC-conjugated sTA (SV40 Tag)–specific mAb (Pab 405).

the opposite flank (Fig. 3A). However, for the group of mice immunized with sTA in alum that rejected a primary s.c. challenge with 4T1.sTA tumor cells, subsequent rejection of a secondary s.c. challenge with parental 4T1 tumor cells given 2 weeks later in the opposite flank was observed (Fig. 3B). These data suggest that mice were protected from mKSA and 4T1.sTA tumor growth in the primary challenge due to the specific immune response that was generated following immunization with sTA in alum as seen in Fig. 2A and C and that protection from secondary challenge with parental 4T1 mammary tumor cells (Fig. 3B) was due to crosspresentation of unknown TAA shared by 4T1.sTA and parental 4T1 tumor cells resulting in determinant spreading and rejection of parental 4T1 tumor cells. This was supported by the observation that rejection of primary challenge with the unrelated kidneyderived mKSA tumor cells, although sTA-based vaccine driven, had no effect on 4T1 mammary-derived tumor cell growth (Fig. 3A). Thus, the initial immune response induced by immunization with


sTA in alum was specific for the model sTA and enabled mice to reject 4T1.sTA tumor challenge resulting in the initiation of determinant spreading and cross-priming to unknown TAA in 4T1 tumor cells, providing protection against secondary s.c. challenge with parental 4T1 tumor cells. In addition to sTA-based, vaccine-induced protection against s.c. challenge with parental 4T1 tumor cells as a result of apparent determinant spreading, immunization with sTA in alum prevented the formation of spontaneous 4T1 lung metastases in mice that rejected a primary s.c. challenge with 4T1.sTA tumor cells. Mice immunized with sTA in alum that rejected a primary s.c. challenge of mKSA tumor cells did not reject the formation of parental 4T1 spontaneous lung metastases following the secondary s.c. challenge with parental 4T1 tumor cells (Table 2; Fig. 3C, c). However, when mice were immunized with sTA in alum and rejected a primary s.c. challenge with 4T1.sTA, a secondary s.c. challenge with parental 4T1 tumor cells was rejected and the formation of spontaneous

2941

Cancer Res 2005; 65: (7). April 1, 2005

Cancer Research

Figure 2. Effect of sTA immunization on tumor challenge. Female BALB/c mice were immunized four times every 7 days with PBS, alum, or sTA in alum and challenged s.c. with either (A) sTA-positive mKSA kidney-derived tumor cells, (B) sTA-negative 4T1 mammary tumor cells, or (C) sTA-positive 4T1.sTA mammary tumor cells. Tumor growth was monitored thrice weekly (n = 5). Representative of three experiments.

non-H-2d (H-2b, C57/b6-derived) tumor cells (b6SVT and TRAMPC1) as depicted in Fig. 3D (inset). Moreover, when the 4T1-specific T cells were cultured in the presence of H-2d–specific mAbs, the secretion of IFN-g was inhibited. When compared with a control Kbspecific mAb (AF6-88.5, BD Biosciences), the Kd mAb (SF1-1.1, BD Biosciences) completely inhibited IFN-g secretion, whereas the Dd mAb (34-2-12, BD Biosciences) and Ld mAb (8F12, BD Biosciences), which also cross-reacts with Db, inhibited f40% and 50%, respectively. A distinct mAb with cross-reactivity for Ld and Db, shown as LdDb in Fig. 3D (28-14-8, BD Biosciences), was capable of f50% inhibition similar to the distinct LdDb mAb depicted as Ld (8F12; Fig. 3D). The inhibition by the two mAbs that cross-react with Db does not likely reflect inhibition of T-cell receptor and MHC interaction but instead CD8 and MHC interaction via binding of the a3 domain of the MHC by the cross-reactive mAbs and therefore is not directly comparable with the inhibition seen by the Kd and Dd mAbs but supports the possible role of CD8+ T cells given the inhibition based on MHC-I and CD8 interaction. Taken together, these data show that a sTA-based vaccine will induce determinant spreading and rejection of indigenous tumors in the absence of known, defined TAA and that H-2d–restricted cytotoxic T cells may play an important role in rejecting parental 4T1 tumor cell challenge via recognition of unknown TAA expressed by parental as well as sTA-modified 4T1 tumor cells. Treatment of established 4T1 tumors as a result of surrogate tumor antigen–induced determinant spreading. Finally, we evaluated whether vaccinating with a model sTA would have an effect on established tumors in a treatment scenario. Mice were inoculated s.c. simultaneously with 4T1 and 4T1.sTA tumor cells in contralateral flanks. Three days after tumor inoculation, animals in each group were immunized i.p. with sTA in alum or alum alone on days 3, 10, 17, and 24. Growth of the tumors on each flank was monitored and measurements were taken with calipers as described in Materials and Methods. Given that it

parental 4T1 lung metastases was prevented (Table 2; Fig. 3C, d). Strikingly, no 4T1 tumor nodules were visible in the lungs of mice immunized with sTA in alum followed by rejection of a primary s.c. challenge with 4T1.sTA tumor cells (Fig. 3C, d). In contrast, parental 4T1 lung tumor nodules ranged from 2 to 17 in mice that rejected a primary s.c. challenge with mKSA tumor cells but failed to reject a subsequent challenge with parental 4T1 tumor cells (Table 2). These results suggest that unknown TAA present in 4T1.sTA cross-primed the immune system to enable rejection of secondary parental 4T1 tumor challenge and spontaneous lung metastasis. To determine whether MHC-I (H-2)–restricted cytotoxic T cells were generated following sTA-induced cross-protection, splenocytes from mice that rejected parental 4T1 tumor challenge were analyzed for cytokine production following tumor cell stimulation in vitro. Splenocytes cultured with irradiated, parental 4T1 tumor cells generated T cells that secreted IFN-g following recognition of 4T1 tumor cells in a standard 24-hour cytokine release assay (Fig. 3D). The T cells seemed to be specific for the BALB/c, H-2d 4T1 tumor cells as they failed to secrete detectable levels of IFN-g when cultured with H-2d-matched, syngeneic tumor cells (mKSA) and

Cancer Res 2005; 65: (7). April 1, 2005

2942

Table 1. Effect of sTA immunization on the number of spontaneous lung metastases following challenge with 4T1.sTA or 4T1 Challenge

sTA antibody titer

4T1.sTA

Yes

4T1

Yes

S.c. tumor

Day 66 0* 0 0 0 Day 42 2,052* 2,138 2,304

No. lung metastases

0c 0 0 0 12c 10 18

NOTE: Mice were immunized i.p. with sTA in alum four times at 1-week intervals and challenged s.c. with tumor cells 14 days following the last immunization. All values represent s.c. tumor growth (mm3) and number of lung metastases for individual mice. *S.c. tumor size was determined with calipers, measuring perpendicular angles. cLungs were harvested when s.c. tumor size reached f1 1 cm or by the day indicated. Tissue was stained with India ink solution and destained with Fekete’s solution and tumor nodules were counted and recorded.



Figure 3. sTA-induced cross-protection is 4T1 tumor specific. A and B, effect of sTA immunization on 4T1 tumor growth following rejection of a primary challenge with either (A ) mKSA tumor cells (sTA-positive control) or (B ) 4T1.sTA tumor cells. Mice were immunized with sTA in alum as described in Materials and Methods and given a primary s.c. challenge with either mKSA or 4T1.sTA followed 2 weeks later by a secondary s.c. challenge in the opposite flank with parental 4T1 tumor cells. 4T1 tumor growth (n = 5). Representative of three experiments. Curves represent s.c. tumor growth in individual mice. C, spontaneous 4T1 lung metastases. a, 4T1 challenge following alum immunization; b, 4T1 challenge following sTA in alum immunization; c, 4T1 challenge following sTA immunization and mKSA tumor rejection; d, 4T1 challenge following sTA immunization and 4T1.sTA rejection. Arrows, examples of tumor nodules. d, lungs are tumor free. Representative of three experiments. D, T-cell response in sTA-induced 4T1 cross-protected mice. Splenocytes from 4T1 cross-protected mice were harvested and cultured 5 to 7 days with irradiated 4T1 tumor cells. Stimulated spleen cells were then cultured for 24 hours at 1:1 with 4T1 tumor cells in the presence or absence of MHC-I-specific antibodies and supernatants were assayed for the presence of IFN-g by ELISA. A mAb with specificity for H-2Kb served as a negative control (cntrl ). Individual columns represent values minus those for T cells alone (i.e., no antibody). Columns, mean of triplicate determinations for individual mice from three separate experiments; bars, SE.

takes f10 to 14 days to induce immunity following immunization, this represents a 13- to 17-day established tumor, albeit minimal disease. Immunization with sTA in alum resulted in the cure of 4T1.sTA tumors in five of five mice (Fig. 4C), whereas alum had no significant effect on 4T1.sTA s.c. tumor growth (data not shown). Mice that were immunized with alum alone also failed to regress 4T1 parental tumors in the opposite flank (Fig. 4A). Strikingly, two of five mice were cured and two additional mice had a significant reduction in tumor growth with delayed onset of their parental 4T1 tumors when immunized with the sTA and simultaneously inoculated with 4T1.sTA and 4T1 tumor cells (Fig. 4B). All five mice in this group were cured of their right flank sTA-expressing 4T1 tumors (Fig. 4C) and 80% (4 of 5) had a complete or partial response against their parental


4T1.sTA-negative left flank tumors (Fig. 4B), demonstrating that it may be possible to treat established disease with sTA-based vaccination to induce determinant spreading to unknown TAA expressed by indigenous tumors, represented herein by parental 4T1 tumor cells.

Discussion No other form of medical intervention has had a greater impact on world health than vaccination. The identification of tumor rejection antigens and the genes that encode them, with the development of relevant animal tumor models involving the very same antigens, should provide the information and opportunity necessary to bring strategies for active immunization against many

2943

Cancer Res 2005; 65: (7). April 1, 2005

Cancer Research

intact tumor cells that share the antigen and the appropriate MHC, recognize the tumor cells, and destroy them. The proper maturation and activation of the dendritic cells is required before they are capable of priming naive T cells. The microenvironment and cytokines necessary and capable of dendritic cell activation are best elicited by infectious agents or danger signals induced by infectious agents. Our model sTA is a viral protein and as such

Table 2. Effect of 4T1.sTA and mKSA rejection following sTA immunization on the formation of spontaneous 4T1 lung metastases Immunization

sTA in alum

sTA in alum

Primary challenge

Secondary challenge

mKSA (s.c.) 1e6 1e6 1e6 1e6 4T1.sTA (s.c.) 1e5 1e5 1e5

4T1 (s.c.) 1e4 1e4 1e4 1e4 4T1 (s.c.) 1e4 1e4 1e4

1e5

1e4

Days after challenge with 4T1

No. 4T1 lung tumors

35 44 44 44

17* 2 16 2

52 52 52

0c 0 0

52

0

NOTE: Mice were immunized i.p. with sTA in alum four times at 1-week intervals and challenged s.c. with tumor cells following the last immunization. All values represent the number of lung metastases for individual mice. *Mice challenged with mKSA following sTA immunization had no effect on subsequent 4T1 tumor growth or the formation of spontaneous 4T1 lung metastases. cMice challenged with 4T1.sTA following immunization rejected parental 4T1 tumor growth and prevented the formation of spontaneous 4T1 lung.

forms of human cancer into the clinic. In the past 5 to 7 years alone, great strides have been made toward the definition of relevant molecular targets on human tumor cells that may serve as TAAs for immunotherapy and vaccine development. Active vaccination against cancer is a treatment approach that has proven effective in preclinical murine models. However, vaccines have been difficult to exploit for highly tumorigenic and poorly immunogenic tumors, such as the 4T1 mammary carcinoma model for which no TAAs have been reported. We hypothesized that in the absence of known TAA it is possible to induce tumor-specific immunity using a vaccine that is composed of a sTA to induce cross-presentation of unknown TAA and tumor-specific immunity as a result of intermolecular determinant spreading. To address this, we used the 4T1 murine mammary carcinoma model and SV40 Tag as a sTA to induce tumor-specific immunity and subsequent cross-presentation of unknown mammary carcinomaassociated antigens. The observation that T cells could be primed against antigens carried by cells lacking the proper MHC necessary to be recognized by T cells was first reported by Bevan (24). Since the first reports, the phenomenon has been extensively studied in models of autoimmunity, infectious disease, and, to a lesser degree, cancer (25). It has been shown that cross-presentation is mediated by professional, bone marrow–derived antigen-presenting cells and most potently by dendritic cells (26). For example, as tumor cells die or are actively destroyed, they release their complement of antigens into the microenvironment. These antigens are captured and processed by dendritic cells, which then migrate to regional lymph nodes to present the processed antigens to naive T cells. Once primed and activated, these T cells are capable of finding

Cancer Res 2005; 65: (7). April 1, 2005

Figure 4. sTA-induced rejection of established 4T1 tumors. Mice were simultaneously challenged s.c. in contralateral flanks with 4T1.sTA and 4T1 tumor cells and immunized on days 3, 10, 17, and 24 with either sTA in alum or alum alone. Tumor development was monitored and recorded. A, treatment of 4T1 tumor cells as a result of immunization with alum alone (n = 4). B, treatment of 4T1 tumor cells as a result of immunization with sTA in alum (n = 5). C, treatment of 4T1.sTA tumor cells as a result of immunization with sTA in alum (n = 5). Representative of two separate experiments.

2944



possesses the immunogenicity required to induce an immunologic environment that would be optimal for antigen crosspresentation and subsequent determinant spreading. Previously, we reported that SV40 Tag was capable of eliciting protective tumor immunity in BALB/c mice (27) and that the primary mechanism of immunity was antibody dependent and CD8+ T cell independent (28). In addition, we showed that CD4+ T cells are important for the induction of Tag-specific tumor protective immune responses (29). Based on these data, we used SV40 Tag as a model sTA to transfect the 4T1 mammary tumor cell line enabling us to immunize against this model sTA, subsequently challenge with the sTA-expressing 4T1 tumor cell line and the parental 4T1 tumor cell line, and assess the induction of protective immunity against parental 4T1 tumor challenge. The subline 4T1.sTA was generated by transfecting the parental cell line 4T1 with a plasmid containing the full-length cDNA for SV40 Tag, the model sTA. The presence of sTA in 4T1.sTA cells had no observable effect on the rate of s.c. tumor formation or spontaneous lung metastasis in naive mice compared with parental 4T1 tumor cells. In the present studies, we immunized naive mice with recombinant SV40 Tag protein as a model sTA, which resulted in the induction of sTA-specific immunity that had no effect on parental 4T1 tumor formation but completely protected against challenge with sTA-transfected 4T1 tumor cells as would be expected. However, 4T1 tumor growth and spontaneous lung metastasis was prevented when mice were challenged with 4T1 tumor cells following rejection of 4T1.sTA tumor cell challenge. This effect was not seen when unrelated sTA-positive tumor cells were rejected before parental 4T1 tumor cell challenge, suggesting the cross-presentation of shared murine mammary carcinoma antigens and the induction of systemic immunity via intermolecular determinant spreading. We went a step further to evaluate the effect of this vaccine approach on established disease and showed that we were able to dramatically slow the growth of both 4T1.sTA and parental 4T1 established tumors following immunization with the sTA and to cure 40% (2 of 5) of the mice of their parental nonsTA-expressing 4T1 tumors. Together, these data suggest that prior vaccination with a sTA expressed as a transgene in poorly immunogenic tumor cells activates cross-presentation of unknown tumor-specific antigens and the induction of systemic tumor immunity via the targeted immunologic destruction of the sTAexpressing tumor cells. Previous studies by other investigators employing the 4T1 tumor model showed reduction in metastases of 4T1 (30–32), prolong survival (30, 32, 33), and a slowing in growth of primary tumors (6, 34); however, few reported cure of the primary tumor or complete protection from spontaneous lung metastasis. Using the sTA-based vaccine, we were able to elicit a potent immune response that resulted in complete protection from parental 4T1 tumor challenge and rejection of the formation of lethal, spontaneous lung metastases. Moreover, we were able to cure 40% of the animals in a treatment model of established disease. Our initial studies indicate that CD8+ T cells may play a role in the rejection of the non-sTA-expressing parental 4T1 tumors. However, additional studies, such as in vivo depletion of T-cell subsets (29), will need to be done to further elucidate the mechanism of rejection. Previously, we reported that antibody-mediated rejection of SV40 Tag–expressing tumors is a primary mechanism following Tag immunization (28); this is possible due to the surface expression of SV40 Tag on SV40-transformed or Tag-transfected cells. Additional studies will be done to confirm that antibodies


mediate the immune destruction of the sTA-expressing tumor, which lead to cross-presentation of unknown TAA and determinant spreading that results in CTL activation and rejection of nonsTA-expressing parental tumors. Evidence to this effect is our recent unpublished observation that 4T1 cells expressing ovalbumin following transfection with ovalbumin plasmid (kind gift from Dr. M.J. Bevan, University of Washington, Seattle, WA) as a model sTA are not rejected in vivo following ovalbumin in alum immunization and the induction of ovalbumin-specific IgG antibodies. This is in contrast to rejection of 4T1 transfected with SV40 Tag as the model sTA as described herein and may reflect the simple lack of surface expression of ovalbumin on 4T1 tumor cells following transfection, thus supporting a role for antibodies in the afferent phase of this vaccination regimen. We have subcloned the ovalbumin cDNA into a construct that was engineered to direct surface expression of the transgene and will use this to transfect 4T1 tumor cells and further evaluate the role of antibody as a mechanism in rejecting sTA-expressing tumors following sTA protein immunization. Others have approached the idea of vaccinating against cancer by using hapten-coated tumor cells or allo-MHC molecules as the foreign antigen (35, 36) called xenogenization (37, 38). Haptens are not as immunogenic as whole foreign proteins and the mechanisms of immunity may be different for allo-MHC than for foreign protein antigens. These approaches largely centered on administration of the modified or xenogenized tumor cells as the vaccine and were minimally effective. Whole cell xenogenized tumor vaccines may not elicit the proper immune response necessary to induce potent tumor immunity via cross-presentation and determinant spreading, which is likely why these approaches have largely been abandoned. A variation on this approach was to mutagenize the nonimmunogenic tumor cells, immunize with the mutant cells, and challenge with parental tumor cells. Van Pel and Boon showed that rejection of a nonimmunogenic thymic leukemia cell line could be accomplished if the challenge followed immunization with the same cell line that was treated with the mutagen N-methyl-NV-nitro-N-nitrosoguanidine (39); this approach led to the discovery of the first human melanoma-associated antigen. Cross-presentation in human cancer vaccine studies has been reported and is believed to be important in rejection of cancer in patients following immunotherapy. Thomas et al. showed the first direct evidence that T-cell responses can be generated via cross-presentation in pancreatic cancer patients vaccinated with pancreatic cancer cell lines engineered to secrete granulocyte macrophage colony-stimulating factor. Three patients generated MHC-I-restricted T-cell responses against epitopes in the TAA mesothelin carried by the vaccine cancer cell lines, although none of the patients possessed the MHC-I of the vaccine cancer cell line (40). In addition, human monocyte-derived dendritic cells were shown to be more effective at cross-presentation in association with patients with head and neck cancer when loaded with apoptotic tumor cells if proinflammatory cytokines or CD40L was included in the preparation (41). Our approach is unique in that it employs the use of a defined, foreign protein as a sTA-based vaccine to drive determinant spreading to unknown TAA. By vaccinating with a nonself sTA, an immune response is induced that focuses the proper immunologic mechanism(s) and environment on the target tumor cells, enabling the potent cross-priming of immune cells for the subsequent rejection of the indigenous/parental tumor.

2945

Cancer Res 2005; 65: (7). April 1, 2005

Cancer Research

Acknowledgments Received 8/9/2004; revised 12/30/2004; accepted 1/28/2005. Grant support: NIH grant CA77351 and Southwest Cancer Treatment and Research Center.

References 1. Bright RK, Jumper CA. Immunotherapy. In: Pass HI, Carbone DP, Johnson DH, Minna JD, Turrisi III AT, editors. Lung cancer: principles and practice. 3rd ed. Philadelphia: Lippincott Williams & Wilkins; 2004. 2. Gilboa E. The promise of cancer vaccines. Nat Rev Cancer 2004;4:401–11. 3. Lewis JD, Reilly BD, Bright RK. Tumor-associated antigens: from discovery to immunity. Int Rev Immunol 2003;22:81–112. 4. Aslakson CJ, Miller FR. Selective events in the metastatic process defined by analysis of the sequential dissemination of subpopulations of a mouse mammary tumor. Cancer Res 1992;52:1399–405. 5. Pulaski B, Ostrand-Rosenberg S. MHC class II and B7.1 immunotherapeutic cell-based vaccine reduces spontaneous mammary carcinoma metastases without affecting primary tumor growth. Cancer Res 1998;58:1486–93. 6. Coveney E, Wheatley GH, Lyerly HK. Active immunization using dendritic cells mixed with tumor cells inhibits the growth of primary breast cancer. Surgery 1997;122:228–34. 7. Pulaski BA, Clements VK, Pipeling MR, OstrandRosenberg S. Immunotherapy with vaccines combining MHC class II/CD80+ tumor cells with interleukin-12 reduces established metastatic disease and stimulates immune effectors and monokine induced interferon g. Cancer Immunol Immunother 2000;49:34–45. 8. Lehmann PV, Forsthuber T, Miller A, Sercarz EE. Spreading of T-cell autoimmunity to cryptic determinants of an autoantigen. Nature 1992;358:155–7. 9. Lehmann PV, Sercarz EE, Forsthuber T, Dayan CM, Gammon G. Determinant spreading and the dynamics of the autoimmune T-cell repertoire. Immunol Today 1993;14:203–8. 10. Ribas A, Timmerman JM, Butterfield LH, Economou JS. Determinant spreading and tumor responses after peptide-based cancer immunotherapy. Trends Immunol 2003;24:58–61. 11. Vanderlugt CL, Miller SD. Epitope spreading in immune-mediated disease: implications for immunotherapy. Nat Rev Immunol 2002;21:85–95. 12. Sercarz EE. Driver clones and determinant spreading. J Autoimmun 2000;14:275–7. 13. El-Shami K, Tirosh B, Bar-Haim E, et al. MHC class I-restricted epitope spreading in the context of tumor rejection following vaccination with a single immunodominant CTL epitope. Eur J Immunol 1999;29: 3295–301. 14. Markiewicz MA, Fallarino F, Ashikari A, Gajewski TF. Epitope spreading upon P815 tumor rejection triggered by vaccination with the single class I MHC-restricted peptide P1A. Int Immunol 2001;13:625–32. 15. Lanford RE. Expression of simian virus 40 T antigen in insect cells using a baculovirus expression vector. Virology 1988;167:72–81.

Cancer Res 2005; 65: (7). April 1, 2005

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

16. Shearer MH, Lanford RL, Kennedy RC. Monoclonal anti-idiotypic antibodies induce humoral immune responses specific for simian virus 40 large tumor antigen in mice. J Immunol 1990;145:932–9. 17. Kit ST, Kurimura T, Dubbs DR. Transplantable mouse tumor line induced by injection of SV40-transformed mouse kidney cells. Int J Cancer 1969;4:384–92. 18. Bright RK, Shearer MH, Pass HI, Kennedy RC. Immunotherapy of SV40-induced tumors in mice: a model for vaccine development. Dev Biol Stand 1998; 94:341–53. 19. Bright RK, Kimchi ET, Shearer MH, Kennedy RC, Pass HI. SV40 tag specific cytotoxic T lymphocytes generated from the peripheral blood of malignant pleural mesothelioma patients. Cancer Immunol Immunother 2002;50:682–90. 20. Bright RK, Shearer MH, Kennedy RC. SV40 large tumor antigen associated synthetic peptides define native antigenic determinants and induce protective tumor immunity in mice. Mol Immunol 1994;14: 1077–87. 21. Bright RK, Shearer MH, Kennedy RC. Fine specificity of the murine immune response to SV40 large tumour antigen utilizing synthetic peptides that define selected epitopes. Clin Exp Immunol 1994;96:491–9. 22. Bright RK, Shearer MH, Kennedy RC. Comparison of the murine humoral immune response to recombinant simian virus 40 large tumor antigen: epitope specificity and idiotype expression. Cancer Immunol Immunother 1993;37:31–9. 23. Shearer MH, Bright RK, Kennedy RC. Comparison of humoral immune responses and tumor immunity in mice immunized with recombinant SV40 large tumor antigen and a monoclonal anti-idiotype. Cancer Res 1993;53:5734–9. 24. Bevan MJ. Cross-priming for a secondary cytotoxic response to minor H antigens with H-2 congenic cells which do not cross-react in the cytotoxic assay. J Exp Med 1976;143:1283–8. 25. Heath WR, Carbone FR. Cross-presentation, dendritic cells, tolerance, and immunity. Annu Rev Immunol 2001;19:47–64. 26. Huang AY, Golumbek P, Ahmadzedah M, Jaffee E, Pardoll D, Levitsky H. Role of bone marrow-derived cells in presenting MHC I-restricted tumor antigens. Science 1994;264:961–5. 27. Bright RK, Beames B, Shearer MH, Kennedy RC. Protection against a lethal challenge with SV40-transformed cells by the direct injection of DNA-encoding SV40 large tumor antigen. Cancer Res 1996;56:1126–30. 28. Bright RK, Shearer MH, Kennedy RC. Immunization of BALB/c mice with recombinant simian virus 40 large tumor antigen induces antibody-dependent cellmediated cytotoxicity against simian virus 40-transformed cells. An antibody based mechanism for tumor immunity. J Immunol 1994;153:2064–71. 29. Kennedy RC, Shearer MH, Watts AM, Bright RK. CD4+ lymphocytes play a critical role in antibody

2946

production and tumor immunity against simian virus 40 large tumor antigen. Cancer Res 2003;63:1040–5. 30. Ostrand-Rosenberg S, Pulaski BA, Armstrong TD, Clements VK. Immunotherapy of established tumor with MHC class II and B7.1 cell-based tumor vaccines. Adv Exp Med Biol 1998;451:259–64. 31. Rakhmilevich AL, Janssen K, Hao Z, Sondel PM, Yang N-S. Interleukin-12 gene therapy of a weakly immunogenic mouse mammary carcinoma results in reduction of spontaneous lung metastases via a T-cell-independent mechanism. Cancer Gene Ther 2000;1:826–38. 32. Prigozhina TB, Gurevitch O, Morecki S, Yahovlev E, Elkin G, Slavin S. Nonmyeloablative allogeneic bone marrow transplantation as immunotherapy for hematologic malignancies and metastatic solid tumors in preclinical models. Exp Hematol 2002;30:89–96. 33. Pulaski BA, Terman DS, Khan S, Muller E, OstrandRosenberg S. Cooperativity of Staphylococcal aureus enterotoxin B superantigen, major histocompatibility complex class II, and DC80 for immunotherapy of advanced spontaneous metastases in a clinically relevant postoperative mouse breast cancer model. Cancer Res 2000;60:2710–5. 34. Ostrand-Rosenberg S, Grusby MJ, Clements VK. STAT6-deficient mice have enhanced tumor immunity to primary and metastatic mammary carcinoma. J Immunol 2000;165:6015–9. 35. Bauminger S, Yachnin S. Unimpeded growth of tumour in hosts pre-immunized with tyrosyl- or dinithrophenyl-coated tumor cells. Br J Cancer 1972; 26:77–83. 36. Nabel GJ, Gordon D, Bishop DK, et al. Immune response in human melanoma after transfer of an allogeneic class I major histocompatibility complex gene with DNA-liposome complexes. Proc Natl Acad Sci U S A 1996;93:15388–93. 37. Kobayashi H, Sendo F, Shirai T Kaji H, Kodama T. Modification in growth of transplantable rat tumors exposed to friend virus. J Natl Cancer Inst 1969;42: 413–9. 38. Kobayashi H, Gotohda E, Hosokawa M, Kodama J. Inhibition of metastasis in rats immunized with xenogenized autologous tumor cells after excision of the primary tumor. J Natl Cancer Inst 1975;54:997–9. 39. Van Pel A, Boon T. Protection against a nonimmunogenic mouse leukemia by an immunogenic variant obtained by mutagenesis. Proc Natl Acad Sci U S A 1982;79:4718–22. 40. Thomas AM, Santarsiero LM, Lutz ER, et al. Mesothelin-specific CD8+ T cell responses provide evidence of in vivo cross-presentation by antigenpresenting cells in vaccinated pancreatic cancer patients. J Exp Med 2004;200:297–306. 41. Hoffmann TK, Meidenbauer N, Muller-Berghaus J, Storkus WJ, Whiteside TL. Proinflammatory cytokines and CD40 ligand enhance cross-presentation and crosspriming capability of human dendritic cells internalizing apoptotic cells. J Immunother 2001;24:163–71.
