- Email: [email protected]
Dendritic Cell–Tumor Cell Fusion and Staphylococcal Enterotoxin B Treatment in a Pancreatic Tumor Model
Journal of Surgical Research 107, 196 –202 (2002) doi:10.1006/jsre.2001.6497
Dendritic Cell–Tumor Cell Fusion and Staphylococcal Enterotoxin B Treatment in a Pancreatic Tumor Model Elizabeth J. McConnell, M.D.,* Latha B. Pathangey,† Cathy S. Madsen,† Sandra J. Gendler, Ph.D.,† and Pinku Mukherjee, Ph.D.† *Department of Surgery and †Department of Biochemistry and Molecular Biology, Mayo Clinic Scottsdale, Scottsdale, Arizona 85259 Submitted for publication November 20, 2001; published online August 29, 2002
INTRODUCTION Background. Surgical resection of pancreatic tumors removes gross disease but not metastases. Adjuvant therapy such as chemotherapy and radiation treatment is of little value in metastatic pancreatic cancer. The hypothesis of this investigation is that specific and effective immunotherapeutic vaccine (dendritic/tumor cell fusion) will activate cytotoxic T lymphocytes (CTLs), leading to the eradication of spontaneous pancreatic cancer. Methods. We have developed a double transgenic mouse model (MET) that forms spontaneous pancreatic tumors and expresses the human MUC1 antigen. Sevenweek-old MET mice (n ⴝ 8) were treated every 3 weeks with the vaccine. In addition, these mice received 50 g of superantigen staphylococcal enterotoxin B (SEB), a known T cell stimulant, prior to the first vaccination. A second treatment group received SEB alone (n ⴝ 8) and controls received no treatment (n ⴝ 9). MUC1specific CTLs were measured by chromium release assay. At 10 weeks of age and at necropsy, MUC1 serum levels were measured using a MUC1-specific ELISA. Results. Mice were known to harbor microscopic foci of cancer at birth. Survival was enhanced in vaccine as well as SEB-treated mice (75% CI ⴞ 0.42) compared to controls (11% CI ⴞ 0.28) and both groups of treated mice exhibited mature CTLs without in vitro stimulation. MUC1 serum levels of the vaccine group were 50% less than that of control (P < 0.04) at 10 weeks. MUC1 serum levels directly correlated with tumor weight at necropsy (r ⴝ 公0.86). Conclusions. This is the first evidence that MUC1specific CTLs can be stimulated to enhance survival in a spontaneous tumor model. © 2002 Elsevier Science (USA) Key Words: mouse model; cytotoxic T lymphocyte; pancreas; spontaneous tumor; dendritic cells; MUC1. Presented at the Annual Meeting of the Association for Academic Surgery, Milwaukee, Wisconsin, November 15–17, 2001.
0022-4804/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.
Advanced pancreatic cancer is uniformly fatal because no effective adjuvant therapy exists. Novel therapeutic approaches are needed to improve patient survival. A potential approach is the development of tumor-specific antigen loaded dendritic cell (DC) based immunizations directed at tumor-specific antigens. Among these antigens is epithelial mucin, MUC1, which is highly overexpressed by pancreatic cancer in humans. MUC1-specific cellular and humoral immune responses have been detected in humans, which makes MUC1 an attractive target antigen for immunotherapy. We have previously described a mouse model of spontaneous pancreatic cancer, designated MET [1–3]. By 7 weeks of age, the pancreatic pathology exhibits dysplastic acinar tissue and invasive cancer. By 10 weeks of age, MUC1 levels are measurable in the serum. These animals develop an identifiable immune response to MUC1 but 90% of mice die of pancreatic cancer by 16 weeks of age. This acinar cell pancreatic cancer in a mouse model closely resembles the progression of human pancreatic cancer. In addition, MUC1 expression as well as MUC1-specific immune responses can be used to follow the effects of immunotherapeutic strategies. For patients with localized pancreatic cancer, treatment such as a total or subtotal pancreatectomy, radiation therapy, and chemotherapy present serious risks, including diabetes, malabsorption syndromes, and liver failure. Moreover, these therapies have not demonstrably decreased mortality rate. A possible adjuvant treatment for patients with pancreatic cancer is immunotherapy. Limited success has been reported from attempts to treat patients with tumors by adoptive transfer of cytotoxic T lymphocytes (CTLs) [4]. To
196
McCONNELL ET AL.: DC–TUMOR FUSION IMMUNOTHERAPY
date, successful treatment is hindered because of poor antigen stimulation of the CTLs, the inability to activate the available immune repertoire, and the accessibility of the antigen in the secondary lymphoid organs where the CTLs function. We have demonstrated the production of effective naturally occurring CTLs in the MET model [2, 3]. The purpose of this study was to develop a long-term CTL response and improved survival with an immunotherapeutic approach. We used MUC1 expression in the serum, which has been shown in this study to correlate with tumor size, as a marker of tumor progression. MATERIALS AND METHODS Mouse model. The MET mouse, which was characterized previously [2], is a double transgenic created by breeding the MUC1 transgenic (MUC1.Tg) mouse [1] with an oncogene-expressing mouse that spontaneously develops tumors of the pancreas (ET mice) [5]. ET mice were kindly provided by Dr. Judith Tevethia and express the first 127 amino acids of SV40 large T antigen driven by the rat elastase promoter. All mice are on the C57BL/6 strain, an inbred strain of mice that has been well characterized for immunologic studies and has a low incidence of spontaneous tumors. MET mice were genotyped by PCR as previously described [2]. Isolation of bone marrow derived dendritic cells. Allogeneic DCs were obtained by culturing bone marrow cells from the long bone hindlegs of untreated nontransgenic FVB or Balb/c mice. Once harvested, these cells were washed and the red blood cells lysed with 1⫻ pharmlyse (Sigma, St. Louis, MO). The isolated bone marrow cells were then counted and divided into petri dishes (1 ⫻ 10 7 cells per dish) and cultured in DMEM with 10% fetal bovine serum, penicillin (50 U/ml) and streptomycin (50 g/ml), 1% glutamax (Sigma), murine GM-CSF (10 ng/ml) (PharMingen, San Diego, CA), and murine IL-4 (10 ng/ml) (Pharmingen) to select the monocyte population. After 5 days of growth the cells were counted, washed, and phenotyped on a Becton–Dickinson FACscan using the Cell Quest program. This provided antigen-presenting cells which were predominantly CD11c-positive immature DCs. Tumor cell– dendritic cell fusion. Pancreatic tumor cells from a 14- to 18-week-old MET mouse were dissected and manually dissociated utilizing a 40-m nylon strainer (Becton–Dickinson, Mountain View, CA). Red blood cells were lysed with 1⫻ pharmlyse and the remaining pellet was washed, counted, and phenotyped by FACs analysis. Cultured allogenic bone marrow cells from FVB or Balb/c mice were isolated as described above. Immature DCs were fused to the tumor cells at a ratio of 5 DCs to 1 tumor cell. The fusion was performed with 25% PEG and 5% DMSO in PBS solution at 37°C using a method described previously [6]. The fused cells were then irradiated with 200 rads using a Linear Accelerator 2100C from Varian Medical Systems, Inc. (Palo Alto, CA). Fusion efficiency was determined using two-color FACs analyses for cytokeratin (tumor cells) and CD11c (DCs). Fusion efficiency varied from 20 to 50%. Immune strategies. Three treatment groups of 7-week-old MET mice are listed in Table 1 and the immunization strategy as well as the time line is schematically presented in Fig. 1. Two groups were injected intraperitoneally with the superantigen staphylococcal enterotoxin B (SEB) (Sigma, St. Louis, MO) (50 g/500 l/mouse). Within 24 h, an intravenous injection of the dendritic/tumor cell fusion (2.0 ⫻ 10 6 cells/200 l/mouse) was given via the tail vein in half of the superantigen treated group. This intravenous injection was then repeated every 3 weeks (Fig. 1). All experimental procedures on mice were conducted according to the IACUC guidelines. Tumor weights. Mice were sacrificed with CO 2 gas and individually weighed, and a necropsy was performed. The entire pancreatic
197
tumor was dissected free of fat and lymph nodes and weighed. The liver, lungs, kidneys, and peritoneum were examined for nodules. The tumor was fixed in methacarn (60% methanol, 30% chloroform, 10% glacial acetic acid) and paraffin embedded, and 5-m sections were cut. The sections were stained with H&E, and immunohistochemical analysis was performed as described previously [2]. Serum MUC1. Serum MUC1 levels were determined using the Truquant BR ELISA assay supplied by Biomira, Inc. (Edmonton, Canada) [2, 7]. Blood for serum analysis was collected from mice when they were 10 weeks old and at necropsy. Serum was separated and stored at ⫺80°C until the ELISA was performed. Cell line. B16 murine melanoma cell line (derived from C57BL/6 strain) expressing MUC1 (B16.MUC1) was kindly provided by Dr. M. A. (Tony) Hollingsworth (The Eppley Institute, University of Nebraska Medical Center, Omaha, NE) [1]. The cell line was maintained at 37°F in 5% CO 2 in DMEM medium with 10% FBS, penicillin (50 U/ml), and streptomycin (50 g/ml), supplemented with 300 g/ml G418. Cells were routinely tested by flow cytometry for the presence of MUC1. CTL assays. Splenocytes from individual MET mice were harvested and passaged through a nylon mesh. The red blood cells within the spleen were lysed using pharmlyse (0.45% ammonium chloride solution) from PharMingen. The target cell line, B16.MUC1, was incubated with IFN-␥ (0.5ng/ml) (PharMingen) overnight, to enhance expression of MHC class I molecules. Determination of CTL activity was performed using a standard 8-h 51Cr-release method [8]. Specific 51Cr-release at 8 h was calculated according to the following formula: (experimental CPM-spontaneous release CPM/maximum release CPM-spontaneous release CPM) ⫻ 100. Spontaneous release in all experiments was less than 15% of maximum release. Flow cytometry. Single cells from the tumors of MET mice and DCs from allogeneic mice were analyzed by two-color immunofluorescence for proteins: CD4, CD8, pan-cytokeratin, MUC1, MAC3, CD11c, and MHC class I (PharMingen). Antibodies to CD11c, B7.1, B7.2, MHC class I were used to determine purity of DCs and antibody to pan-cytokeratin was used to identify tumor cells. Flow cytometric analysis was done on Becton–Dickinson FACs using the Cell Quest program. Statistical methods. MUC1 serum levels were compared among three treatment groups (SEB/fusion, SEB, control) at week 10 and at necropsy. Tumor size was also compared among the three groups at necropsy. Statistical significance was assessed using one-way ANOVA model with the Tukey–Kramer adjustment for multiple comparison. The margins of error for the comparisons were obtained by calculating the 95% confidence intervals for the differences between group means. The relationship between MUC1 serum levels and tumor size at necropsy was measured using the Pearson correlation coefficient. Computations were done with SAS/STAT software version 8.2, utilizing the GLM procedure and the correlation procedure (SAS/STAT version 8.2, SAS Institute, Inc., Cary, NC).
RESULTS
Preparation of DC/Fusion Vaccine MET mice 7 weeks of age were grouped into control or treatment groups (Table 1). The treatment groups were SEB superantigen alone (one dose 50 g), or SEB superantigen (one dose) immediately followed by a tumor/DC fusion, which was administered every 3 weeks until day 129 when all mice were sacrificed (Fig. 1). Fusion efficiency was determined using FACs analysis to determine whether an adequate number of fused tumor/DCs was injected (Fig. 2). By FACs analysis, 70% of tumor cells which are epithelial in origin
198
JOURNAL OF SURGICAL RESEARCH: VOL. 107, NO. 2, OCTOBER 2002
TABLE 1 Treatment Groups Treatment groups Control SEB SEB ⫹ fusion
No. of mice
SEB 50 g/500 l (ip)
DCs fused to MET tumor cells 2.0 ⫻ 10 6 cells/200 l PBS (iv)
9 8 8
Injected on day 49 Injected on day 49
Injected on days 50, 71, 92, and 104
express cytokeratin (Fig. 2, column 1). The predominant population of antigen-presenting cells expresses the receptor CD11c, a marker for DCs (Fig. 2, column 2). MHC class I and II are present on both the tumor and the DCs (data not shown). Few of the DCs expressed cytokeratin and none of the tumor cells expressed CD11c (Fig. 2). The fusion efficiency was determined by two-color immunofluorescence staining, using cytokeratin and CD11c antibodies specific for the tumor cell and the DC, respectively. The fusion efficiency varied from 20 to 50% for each fusion (all fusion data not shown). MET Mice Survival Enhanced with Treatment All mice were allowed to survive at least 129 days. All mice in the control group had died by 129 days and the necessary follow-up time for the other groups was not known and may have been very long. Kaplan– Meier analysis compares survival across the whole curve. Figure 3 shows a significant difference (P ⬍ 0.01) in survival between the group treated with SEB/DC fusion or SEB alone (75% CI ⫾ 0.42) compared to control (11% CI ⫾ 0.28). MUC1 Serum Levels Correlate with Tumor Size During the experiment, MUC1 serum levels were measured at 10 weeks of age and necropsy (Figs. 4A and 4B). There was a significant difference between the SEB/fusion group and the control group at 10 weeks of
age (P ⫽ 0.04) (Fig. 4A). The sample size was too small to assess a difference in serum levels between the SEB group and the control group. At necropsy there was no difference in serum levels (Fig. 4B). However, MUC1 serum levels acquired at necropsy correlated significantly with tumor size (Pearson correlation coefficient r ⫽ 公0.86) (Fig. 4C). Immunization Elicits Mature CTLs At necropsy, CTLs were obtained from splenocytes and tested for reactivity against MUC1 expressing B16 melanoma tumor cells in a standard 8-h Cr-release assay [2, 8]. The mice treated with SEB/DC fusion as well as with SEB exhibited high levels of the MUC1specific CTLs (Fig. 5). The CTL activity in control mice, which were sacrificed at morbidity, is shown in the figure but was not included in the analysis, as CTL activity diminishes greatly at morbidity. The CTLs were detected in freshly isolated splenocytes without any in vitro manipulation, suggesting that the CTLs were mature in nature, whereas the naturally occurring CTLs in MET mice described previously were precursor CTLs [3]. Although the CTLs were cytotoxic in vitro, they apparently had little effect on the growing tumors in vivo in the MET mice. Tumors from both treated and control groups were examined using immunohistochemistry. We showed previously that the dysplastic acinar cells that had developed by 3 weeks of age expressed underglycosy-
FIG. 1. The immunization strategy and time line showing tumor progression. There was microscopic growth of pancreatic tumor when treatment began. Time points of immunologic monitoring are shown.
McCONNELL ET AL.: DC–TUMOR FUSION IMMUNOTHERAPY
199
FIG. 2. Characterization of the fusion vaccine. Flow cytometric analysis of antigens of tumor cells and DCs is represented. MET tumor cells were labeled with antibody bearing fluorescent dyes to evaluate the expression and presence of cytokeratin. Antibody to CD11c is the prominent antigen expressed on DCs. Fusion of the DCs and MET tumor was at a ratio of 5:1. DCs fused with MET tumor cells (DC:tumor fusion) were analyzed by flow cytometry. The percentage of cells expressing both cytokeratin and CD11c represents the fused cells.
lated MUC1 as determined by reactivity with the monoclonal antibodies HMFG-2, SM-3, and BC2 [2]. A similar staining pattern was observed by all three groups of mice
FIG. 3. The Kaplan–Meier survival of the mice after treatment. The Kaplan–Meier survival curve demonstrated improved survival of SEB/fusion (E) and SEB (‚) treatment mice (P ⬍ 0.01) based on a Fisher exact test.
in this study, regardless of the treatment strategy. Tumors progressed to well-differentiated acinar cell carcinomas by 15 weeks and continued to express MUC1 strongly by immunohistochemistry (data not shown). Additionally, both treated and control MET mice developed large solid tumor masses of less well-differentiated acinar cell carcinomas which failed to express human MUC1 strongly, although the cytoplasmic tail monoclonal antibody CT2, which is glycosylation insensitive and reacts with both human and mouse Muc1, showed high levels of MUC1 cytoplasmic tail expression (data not shown). There is no effective way to differentiate between the mouse and human forms of MUC1 using the cytoplasmic tail monoclonal antibody. In both treated and control mice, there were regions of MUC1 expressing tumor cells and regions of the tumor in which the tandem repeat epitopes of MUC1 appeared to be masked by glycosylation or the MUC1-expressing cells had been eliminated by the CTLs. At this time we have been unable to determine whether MUC1 epitopes have been masked by glycosylation or whether the MUC1 expressing cells have been largely eliminated. Importantly, there was
200
JOURNAL OF SURGICAL RESEARCH: VOL. 107, NO. 2, OCTOBER 2002
FIG. 5. Mature MUC1-specific CTLs were elicited by SEB/fusion and SEB treatments. CTL activity was determined with an 8-h 51 Cr-release assay using B16.MUC1 as tumor target cells. The effector:target ratio was 100:1. Both SEB/fusion and SEB mice showed high levels of mature CTL activity, without any in vitro manipulation. The low level of CTL activity in control mice was probably due to their morbid state.
mice, despite the presence of activated CTLs and improved survival. DISCUSSION
FIG. 4. MUC1 serum levels. MUC1 serum levels were determined by ELISA at 10 weeks of age and at necropsy. A. Demonstration of the correlation with SEB/fusion (E) and significantly lower MUC1 serum levels (P ⬍ 0.04) from a one-way ANOVA model with Tukey–Kramer adjustment for multiple comparisons. B. The MUC1 serum levels are not significantly different between treated and control groups at morbidity. C. The MUC1 serum levels correlate directly to the tumor size (r value ⫽ 公0.86 by Pearson correlation).
A major goal of immunotherapy is to generate tumorspecific CTLs that can effectively eliminate the growing tumor. Previous research has been focused on identifying and characterizing proteins expressed on tumor cells that may serve as potential tumor-specific antigens for recognition by CTLs [9]. Some of the most promising candidates represent conventional cellular proteins that are expressed in both normal and transformed cells [10 –21]. One such candidate is MUC1 [22, 23]. Although MUC1 is normally expressed in epithelial cells lining ducts and glands at low levels, it is a
no detectable difference in the treated groups from the control group. Tumor Size Was Not Altered with Treatment Although there was a trend toward smaller pancreatic tumor size in the treatment groups, the results did not reach statistical significance (Fig. 6). Mice were weighed and examined on a weekly basis. At necropsy, MET mice were rapidly losing weight, an indication of morbidity. Tumors were harvested and wet weight was determined. Tumors progressed rapidly in treatment
FIG. 6. Pancreas weights. Pancreas tumor weights are shown as a percentage of total body weight. Mice were sacrificed and pancreatic tumors were fixed in methacarn, air dried, and weighed. The percentage of tumor weight relative to body weight appeared larger in the control mice but was not found to be statistically significant (P ⫽ 0.67).
McCONNELL ET AL.: DC–TUMOR FUSION IMMUNOTHERAPY
target for immunotherapy, because during tumorigenesis, MUC1 is significantly altered in expression. There is an increase in the amount of MUC1 expressed on cells and in the circulation. The distribution of MUC1 is no longer restricted to the apical surface of the ducts and glands, but is found throughout the tumor mass and on the surface of tumor cells. In addition, the glycosylation of MUC1 is altered in tumor cells [24 – 26]. The oligosaccharides are shorter and fewer in number, leading to the exposure of immunodominant peptide sequences that on normal cell surfaces would be sequestered by glycosylation. The presence of MUC1 on the pancreatic tumor cells allowed us to follow the response to MUC1 following vaccination with a multivalent vaccine approach, DCs fused with tumor cells. In a previous study, we have shown that the MET mice have naturally occurring MUC1-specific precursor CTLs (pCTLs) and that the pCTLs are class I restricted and recognize the tandom repeat region of MUC1 [2, 3]. Importantly, we have shown that the CTLs can be isolated and can eradicate MUC1expressing tumor cells when adoptively transferred in vivo [3]. Although the pCTLs were present in the MET mice, the mice were not protected from the tumor, suggesting that the pCTLs were not activated enough to kill the growing tumor or that the pCTLs arose too late in tumor progression. These data suggested that immunotherapeutic regimens which stimulate MUC1specific pCTLs to become mature may be effective in decreasing tumor burden as well as enhancing survival. In addition, several investigators have shown that primary tumors treated early in their course are susceptible to immune intervention. This has relevance to several human tumors, as class I restricted CTLs have been described in humans in association with other epithelial cancers [27, 28]. In the present study, we hypothesized that specific and effective immunotherapeutic vaccines would activate the naturally existing pCTLs described previously [2] and lead to eradication of pancreatic tumor in the MET mice. Our results demonstrate that: (1) survival was significantly enhanced in vaccine as well as SEB treated MET mice (75%) compared to untreated controls (11%); (2) there was significant increase in mature CTLs in the vaccine treated as well as SEB treated MET mice; (3) MUC1 serum levels at 10 weeks of age in the vaccine treated MET mice were 50% less than that of SEB treated mice or untreated controls (P ⬍ 0.04); and (4) MUC1 serum levels significantly correlated with tumor weight at necropsy. Immunization with SEB alone can activate the pCTLs and may provide substantial help by activating other T cell repertoires needed for effective anti-tumor immune response. There was a significant survival benefit (Fig. 3) with both treatment groups and large numbers of mature CTLs were detected in immunized
201
MET mice. It is possible that the production by tumor cells of TGF, which we showed previously [2], or other immunosuppressive proteins may, in fact, suppress the antitumor effects of the mature CTLs (reviewed in [29]). In addition, it is plausible that within the tumor environment, the mature CTLs that we detected are tolerized or anergized. Understanding the mechanisms of tumor evasion is beyond the scope of this paper and more experiments are necessary to further advance this study. Our results showed a significant decrease in MUC1 serum levels early in tumor progression (10 weeks of age, when tumor burden is low) in mice treated with SEB/fusion (Fig. 4A), suggesting that the vaccination with the fused cells has an effect on tumor growth. Results also clearly indicated a direct correlation between serum MUC1 levels and tumor size (Fig. 4C). This decrease in MUC1 levels with a specific treatment (the fusion vaccine vs SEB, which is a nonspecific immune stimulant) suggests a potential clinical benefit with this approach. Clinically in humans, MUC1 serum levels have been used to track tumor progression [7, 30]. In this mouse model the difference in serum MUC1 levels or tumor weight was undetectable at necropsy (Figs. 4B and 6). One can speculate that this early benefit is lost as this aggressive oncogeneinduced tumor progresses. This is the first evidence demonstrating that mature MUC1-specific CTLs can be elicited by DC/tumor cell fusion and SEB in a spontaneous mouse model of pancreatic cancer. These CTLs were detected prior to clinical evidence of morbidity identified by rapid and acute weight loss, poor grooming, and hunched faces. Moreover, no detectable autoimmune destruction of normal tissues was observed. By following the CTL response to a tumor antigen, in our case MUC1, we should get an estimate of the general tumor-specific CTL response. We did not determine whether there was activity of the infiltrating lymphocytes within the tumors. Data from another spontaneous mouse tumor model suggest that, in the tumor environment, the CTLs may be tolerized (Mukherjee, personal communication). Future studies will address this issue in greater detail. These MET mice appropriately mimic conditions that occur in human tumors and are an excellent model for testing anti-tumor vaccine strategies. In humans, pancreatic cancer is typically found at a late stage of disease when metastatic tumor already exists. To provide an effective immune response after surgical resection of the primary tumor and prevent progression to metastatic tumor growth may be sufficient to enhance survival. Our immunization strategy does not arrest tumor growth entirely, but demonstrates a clear survival benefit in the mice. Future studies into understanding the immunosuppressive effects of these tu-
202
JOURNAL OF SURGICAL RESEARCH: VOL. 107, NO. 2, OCTOBER 2002
mors and tolerance may provide insight into designing effective immunotherapeutic strategies. 15.
ACKNOWLEDGMENTS We are grateful to Biomira, Inc., for supplying reagents and to the surgical pathologists, Drs. C. R. Conley and T. K. Lidner, for analysis of tumor samples. We thank Marvin H. Ruona for computer graphics, Jim Tarara for FACs assistance, and Carol Williams for administrative assistance and manuscript preparation.
16.
17.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13. 14.
Rowse, G. J., Tempero, R. M., VanLith, M. L., Hollingsworth, M. A., and Gendler, S. J. Tolerance and immunity to MUC1 in a human MUC1 transgenic murine model. Cancer Res. 58: 315, 1998. Mukherjee, P., Ginardi, A. R., Madsen, C. S., Sterner, C. J., Adriance, M. C., Tevethia, M. J., and Gendler, S. J. Mice with spontaneous pancreatic cancer naturally develop MUC1specific CTLs that eradicate tumors when adoptively transferred. J. Immunol. 165(6): 3451, 2000. Mukherjee, P., Ginardi, A. R., Tinder, T. L., Sterner, C. J., and Gendler, S. J. MUC1-specific CTLs eradicate tumors when adoptively transferred in vivo. Clin. Cancer Res. 7: 848s, 2001. Rosenberg, S. A., et al. Treatment of patients with metastatic melanoma with autologous tumor-infiltrating lymphocytes and interleukin 2. J. Natl. Cancer Inst. 86(15): 1159, 1994. Tevethia, M. J., Bonneau, R. H., Griffith, J. W., and Mylin, L. A simian virus 40 large T-antigen segment containing amino acids 1 to 127 and expressed under the control of the rat elastase-1 promoter produces pancreatic acinar carcinomas in transgenic mice. J. Virol. 71(11): 8157, 1997. Gong, J., Chen, D., Kashiwaba, M., and Kufe, D. Induction of antitumor activity by immunization with fusions of dendritic and carcinoma cells. Nat. Med. 3(5): 558, 1997. Reddish, M. A., MacLean, G. D., Poppema, S., Berg, A., and Longenecker, B. M. Pre-immunotherapy serum CA27.29 (MUC-1) mucin level and CD69⫹ lymphocytes correlate with effects of Theratope sialyl-Tn-KLH cancer vaccine in active specific immunotherapy. Cancer Immunol. Immunother. 42(5): 303, 1996. Hoffman, R. W., Quinones, R. R., and Shaw, S. Anti-T3 antibody both activates and inhibits the cytotoxic activity of human T cell clones. Behring Inst. Mitt. 77: 30, 1985. Morgan, D. J., Kreuwel, H. T., Fleck, S., Levitsky, H. I., Pardoll, D. M., and Sherman, L. A. Activation of low avidity CTL specific for a self epitope results in tumor rejection but not autoimmunity. J. Immunol. 160(2): 643, 1998. Cabaniols, J. P., Cibotti, R., Kourilsky, P., Kosmatopoulos, K., and Kanellopoulos, J. M. Dose-dependent T cell tolerance to an immunodominant self peptide. Eur. J. Immunol. 24(8): 1743, 1994. Goverman, J., Woods, A., Larson, L., Weiner, L. P., Hood, L., and Zaller, D. M. Transgenic mice that express a myclin basic protein-specific T cell receptor develop spontaneous autoimmunity. Cell 72(4): 551, 1993. von Herrath, M. G., Dockter, J., and Oldstone, M. B. How virus induces a rapid or slow onset insulin-dependent diabetes mellitus in a transgenic model. Immunity 1(3): 231, 1994. Miller, J. F., and Morahan, G. Peripheral T cell tolerance. Annu. Rev. Immunol. 10: 51, 1992. Oehen, S. U., Ohashi, P. S., Burki, K., Hengartner, H., Zinkernagel, R. M., and Aichele, P. Escape of thymocytes and mature
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29. 30.
T cells from clonal deletion due to limiting tolerogen expression levels. Cell. Immunol. 158(2): 342, 1994. Ohashi, P. S., Oehen, S., Buerki, K., Pircher, H., Ohashi, C. T., Odermatt, B., Malissen, B., Zinkernagel, R. M., and Hengartner, H. Ablation of “tolerance” and induction of diabetes by virus infection in viral antigen transgenic mice. Cell 65(2): 305, 1991. Oldstone, M. B., Nerenberg, M., Southern, P., Price, J., and Lewicki, H. Virus infection triggers insulin-dependent diabetes mellitus in a transgenic model: Role of anti-self (virus) immune response. Cell. 65(2): 319, 1991. Poindexter, N. J., Landon, C., Whiteley, P. J., and Kapp, J. A. Comparison of the T cell receptors on insulin-specific hybridomas from insulin transgenic and nontransgenic mice. Loss of a subpopulation of self-reactive clones. J. Immunol. 149(1): 38, 1992. Poplonski, L., Vukusic, B., Pawling, J., Clapoff, S., Roder, J., Hozumi, N., and Wither, J. Tolerance is overcome in beef insulin-transgenic mice by activation of low-affinity autoreactive T cells. Eur. J. Immunol. 26(3): 601, 1996. Whiteley, P. J., Poindexter, N. J., Landon, C., and Kapp, J. A. A peripheral mechanism preserves self-tolerance to a secreted protein in transgenic mice. J. Immunol. 145(5): 1376, 1990. Yule, T. D., Basten, A., and Allen, P. M. Hen egg-white lysozyme-specific T cells elicited in hen egg-white lysozymetransgenic mice retain an imprint of self-tolerance. J. Immunol. 151(6): 3057, 1993. Morgan, D. J., Kreuwel, H. T., and Sherman, L. A. Antigen concentration and precursor frequency determine the rate of CD8⫹ T cell tolerance to peripherally expressed antigens. J. Immunol. 163(2): 723, 1999. Barnd, D. L., Lan, M. S., Metzgar, R. S., and Finn, O. J. Specific, major histocompatibility complex-unrestricted recognition of tumor-associated mucins by human cytotoxic T cells. Proc. Natl. Acad. Sci. USA 86(18): 7159, 1989. Ioannides, C. G., Fisk, B., Jerome, K. R., Irimura, T., Wharton, J. T., and Finn, O. J. Cytotoxic T cells from ovarian malignant tumors can recognize polymorphic epithelial mucin core peptides. J. Immunol. 151(7): 3693, 1993. Burchell, J., Gendler, S., Taylor-Papadimitriou, J., Girling, A., Lewis, A., Millis, R., and Lamport, D. Development and characterization of breast cancer reactive monoclonal antibodies directed against to the core protein of the human milk mucin. Cancer Res. 47: 5476, 1987. Girling, A., Bartkova, J., Burchell, J., Gendler, S., Gillet, C., and Taylor-Papadimitriou, J. A core protein epitope of the polymorphic epithelial mucin detected by the monoclonal antibody SM-3 is selectively exposed in a range of primary carcinomas. Int. J. Cancer 43: 1072, 1989. Burchell, J., and Taylor-Papadimitriou, J. Effect of modification of carbohydrate side chains on the reactivity of antibodies with core-protein epitopes of the MUC1 gene product. Epithelial Cell Biol. 2: 155, 1993. MacLean, G. D., Reddish, M. A., Bowen-Yacyshyn, M. B., Poppema, S., and Longenecker, B. M. Active specific immunotherapy against adenocarcinomas. Cancer Invest. 12(1): 46, 1994. Domenech, N., Henderson, R. A., and Finn, O. J. Identification of an HLA-A11-restricted epitope from the tandem repeat domain of the epithelial tumor antigen mucin. J. Immunol. 155(10): 4766, 1995. Gorelik, L., and Flavell, R. A. Transforming growth factor-beta in T-cell biology. Nat. Rev. Immunol. 2(1): 46, 2002. MacLean, G. D., Reddish, M. A., and Longenecker, B. M. Prognostic significance of preimmunotherapy serum CA27.29 (MUC-1) mucin level after active specific immunotherapy of metastatic adenocarcinoma patients. J. Immunother. 20(1): 70, 1997.
Dendritic Cell–Tumor Cell Fusion and Staphylococcal Enterotoxin B Treatment in a Pancreatic Tumor Model Elizabeth J. McConnell, M.D.,* Latha B. Pathangey,† Cathy S. Madsen,† Sandra J. Gendler, Ph.D.,† and Pinku Mukherjee, Ph.D.† *Department of Surgery and †Department of Biochemistry and Molecular Biology, Mayo Clinic Scottsdale, Scottsdale, Arizona 85259 Submitted for publication November 20, 2001; published online August 29, 2002
INTRODUCTION Background. Surgical resection of pancreatic tumors removes gross disease but not metastases. Adjuvant therapy such as chemotherapy and radiation treatment is of little value in metastatic pancreatic cancer. The hypothesis of this investigation is that specific and effective immunotherapeutic vaccine (dendritic/tumor cell fusion) will activate cytotoxic T lymphocytes (CTLs), leading to the eradication of spontaneous pancreatic cancer. Methods. We have developed a double transgenic mouse model (MET) that forms spontaneous pancreatic tumors and expresses the human MUC1 antigen. Sevenweek-old MET mice (n ⴝ 8) were treated every 3 weeks with the vaccine. In addition, these mice received 50 g of superantigen staphylococcal enterotoxin B (SEB), a known T cell stimulant, prior to the first vaccination. A second treatment group received SEB alone (n ⴝ 8) and controls received no treatment (n ⴝ 9). MUC1specific CTLs were measured by chromium release assay. At 10 weeks of age and at necropsy, MUC1 serum levels were measured using a MUC1-specific ELISA. Results. Mice were known to harbor microscopic foci of cancer at birth. Survival was enhanced in vaccine as well as SEB-treated mice (75% CI ⴞ 0.42) compared to controls (11% CI ⴞ 0.28) and both groups of treated mice exhibited mature CTLs without in vitro stimulation. MUC1 serum levels of the vaccine group were 50% less than that of control (P < 0.04) at 10 weeks. MUC1 serum levels directly correlated with tumor weight at necropsy (r ⴝ 公0.86). Conclusions. This is the first evidence that MUC1specific CTLs can be stimulated to enhance survival in a spontaneous tumor model. © 2002 Elsevier Science (USA) Key Words: mouse model; cytotoxic T lymphocyte; pancreas; spontaneous tumor; dendritic cells; MUC1. Presented at the Annual Meeting of the Association for Academic Surgery, Milwaukee, Wisconsin, November 15–17, 2001.
0022-4804/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.
Advanced pancreatic cancer is uniformly fatal because no effective adjuvant therapy exists. Novel therapeutic approaches are needed to improve patient survival. A potential approach is the development of tumor-specific antigen loaded dendritic cell (DC) based immunizations directed at tumor-specific antigens. Among these antigens is epithelial mucin, MUC1, which is highly overexpressed by pancreatic cancer in humans. MUC1-specific cellular and humoral immune responses have been detected in humans, which makes MUC1 an attractive target antigen for immunotherapy. We have previously described a mouse model of spontaneous pancreatic cancer, designated MET [1–3]. By 7 weeks of age, the pancreatic pathology exhibits dysplastic acinar tissue and invasive cancer. By 10 weeks of age, MUC1 levels are measurable in the serum. These animals develop an identifiable immune response to MUC1 but 90% of mice die of pancreatic cancer by 16 weeks of age. This acinar cell pancreatic cancer in a mouse model closely resembles the progression of human pancreatic cancer. In addition, MUC1 expression as well as MUC1-specific immune responses can be used to follow the effects of immunotherapeutic strategies. For patients with localized pancreatic cancer, treatment such as a total or subtotal pancreatectomy, radiation therapy, and chemotherapy present serious risks, including diabetes, malabsorption syndromes, and liver failure. Moreover, these therapies have not demonstrably decreased mortality rate. A possible adjuvant treatment for patients with pancreatic cancer is immunotherapy. Limited success has been reported from attempts to treat patients with tumors by adoptive transfer of cytotoxic T lymphocytes (CTLs) [4]. To
196
McCONNELL ET AL.: DC–TUMOR FUSION IMMUNOTHERAPY
date, successful treatment is hindered because of poor antigen stimulation of the CTLs, the inability to activate the available immune repertoire, and the accessibility of the antigen in the secondary lymphoid organs where the CTLs function. We have demonstrated the production of effective naturally occurring CTLs in the MET model [2, 3]. The purpose of this study was to develop a long-term CTL response and improved survival with an immunotherapeutic approach. We used MUC1 expression in the serum, which has been shown in this study to correlate with tumor size, as a marker of tumor progression. MATERIALS AND METHODS Mouse model. The MET mouse, which was characterized previously [2], is a double transgenic created by breeding the MUC1 transgenic (MUC1.Tg) mouse [1] with an oncogene-expressing mouse that spontaneously develops tumors of the pancreas (ET mice) [5]. ET mice were kindly provided by Dr. Judith Tevethia and express the first 127 amino acids of SV40 large T antigen driven by the rat elastase promoter. All mice are on the C57BL/6 strain, an inbred strain of mice that has been well characterized for immunologic studies and has a low incidence of spontaneous tumors. MET mice were genotyped by PCR as previously described [2]. Isolation of bone marrow derived dendritic cells. Allogeneic DCs were obtained by culturing bone marrow cells from the long bone hindlegs of untreated nontransgenic FVB or Balb/c mice. Once harvested, these cells were washed and the red blood cells lysed with 1⫻ pharmlyse (Sigma, St. Louis, MO). The isolated bone marrow cells were then counted and divided into petri dishes (1 ⫻ 10 7 cells per dish) and cultured in DMEM with 10% fetal bovine serum, penicillin (50 U/ml) and streptomycin (50 g/ml), 1% glutamax (Sigma), murine GM-CSF (10 ng/ml) (PharMingen, San Diego, CA), and murine IL-4 (10 ng/ml) (Pharmingen) to select the monocyte population. After 5 days of growth the cells were counted, washed, and phenotyped on a Becton–Dickinson FACscan using the Cell Quest program. This provided antigen-presenting cells which were predominantly CD11c-positive immature DCs. Tumor cell– dendritic cell fusion. Pancreatic tumor cells from a 14- to 18-week-old MET mouse were dissected and manually dissociated utilizing a 40-m nylon strainer (Becton–Dickinson, Mountain View, CA). Red blood cells were lysed with 1⫻ pharmlyse and the remaining pellet was washed, counted, and phenotyped by FACs analysis. Cultured allogenic bone marrow cells from FVB or Balb/c mice were isolated as described above. Immature DCs were fused to the tumor cells at a ratio of 5 DCs to 1 tumor cell. The fusion was performed with 25% PEG and 5% DMSO in PBS solution at 37°C using a method described previously [6]. The fused cells were then irradiated with 200 rads using a Linear Accelerator 2100C from Varian Medical Systems, Inc. (Palo Alto, CA). Fusion efficiency was determined using two-color FACs analyses for cytokeratin (tumor cells) and CD11c (DCs). Fusion efficiency varied from 20 to 50%. Immune strategies. Three treatment groups of 7-week-old MET mice are listed in Table 1 and the immunization strategy as well as the time line is schematically presented in Fig. 1. Two groups were injected intraperitoneally with the superantigen staphylococcal enterotoxin B (SEB) (Sigma, St. Louis, MO) (50 g/500 l/mouse). Within 24 h, an intravenous injection of the dendritic/tumor cell fusion (2.0 ⫻ 10 6 cells/200 l/mouse) was given via the tail vein in half of the superantigen treated group. This intravenous injection was then repeated every 3 weeks (Fig. 1). All experimental procedures on mice were conducted according to the IACUC guidelines. Tumor weights. Mice were sacrificed with CO 2 gas and individually weighed, and a necropsy was performed. The entire pancreatic
197
tumor was dissected free of fat and lymph nodes and weighed. The liver, lungs, kidneys, and peritoneum were examined for nodules. The tumor was fixed in methacarn (60% methanol, 30% chloroform, 10% glacial acetic acid) and paraffin embedded, and 5-m sections were cut. The sections were stained with H&E, and immunohistochemical analysis was performed as described previously [2]. Serum MUC1. Serum MUC1 levels were determined using the Truquant BR ELISA assay supplied by Biomira, Inc. (Edmonton, Canada) [2, 7]. Blood for serum analysis was collected from mice when they were 10 weeks old and at necropsy. Serum was separated and stored at ⫺80°C until the ELISA was performed. Cell line. B16 murine melanoma cell line (derived from C57BL/6 strain) expressing MUC1 (B16.MUC1) was kindly provided by Dr. M. A. (Tony) Hollingsworth (The Eppley Institute, University of Nebraska Medical Center, Omaha, NE) [1]. The cell line was maintained at 37°F in 5% CO 2 in DMEM medium with 10% FBS, penicillin (50 U/ml), and streptomycin (50 g/ml), supplemented with 300 g/ml G418. Cells were routinely tested by flow cytometry for the presence of MUC1. CTL assays. Splenocytes from individual MET mice were harvested and passaged through a nylon mesh. The red blood cells within the spleen were lysed using pharmlyse (0.45% ammonium chloride solution) from PharMingen. The target cell line, B16.MUC1, was incubated with IFN-␥ (0.5ng/ml) (PharMingen) overnight, to enhance expression of MHC class I molecules. Determination of CTL activity was performed using a standard 8-h 51Cr-release method [8]. Specific 51Cr-release at 8 h was calculated according to the following formula: (experimental CPM-spontaneous release CPM/maximum release CPM-spontaneous release CPM) ⫻ 100. Spontaneous release in all experiments was less than 15% of maximum release. Flow cytometry. Single cells from the tumors of MET mice and DCs from allogeneic mice were analyzed by two-color immunofluorescence for proteins: CD4, CD8, pan-cytokeratin, MUC1, MAC3, CD11c, and MHC class I (PharMingen). Antibodies to CD11c, B7.1, B7.2, MHC class I were used to determine purity of DCs and antibody to pan-cytokeratin was used to identify tumor cells. Flow cytometric analysis was done on Becton–Dickinson FACs using the Cell Quest program. Statistical methods. MUC1 serum levels were compared among three treatment groups (SEB/fusion, SEB, control) at week 10 and at necropsy. Tumor size was also compared among the three groups at necropsy. Statistical significance was assessed using one-way ANOVA model with the Tukey–Kramer adjustment for multiple comparison. The margins of error for the comparisons were obtained by calculating the 95% confidence intervals for the differences between group means. The relationship between MUC1 serum levels and tumor size at necropsy was measured using the Pearson correlation coefficient. Computations were done with SAS/STAT software version 8.2, utilizing the GLM procedure and the correlation procedure (SAS/STAT version 8.2, SAS Institute, Inc., Cary, NC).
RESULTS
Preparation of DC/Fusion Vaccine MET mice 7 weeks of age were grouped into control or treatment groups (Table 1). The treatment groups were SEB superantigen alone (one dose 50 g), or SEB superantigen (one dose) immediately followed by a tumor/DC fusion, which was administered every 3 weeks until day 129 when all mice were sacrificed (Fig. 1). Fusion efficiency was determined using FACs analysis to determine whether an adequate number of fused tumor/DCs was injected (Fig. 2). By FACs analysis, 70% of tumor cells which are epithelial in origin
198
JOURNAL OF SURGICAL RESEARCH: VOL. 107, NO. 2, OCTOBER 2002
TABLE 1 Treatment Groups Treatment groups Control SEB SEB ⫹ fusion
No. of mice
SEB 50 g/500 l (ip)
DCs fused to MET tumor cells 2.0 ⫻ 10 6 cells/200 l PBS (iv)
9 8 8
Injected on day 49 Injected on day 49
Injected on days 50, 71, 92, and 104
express cytokeratin (Fig. 2, column 1). The predominant population of antigen-presenting cells expresses the receptor CD11c, a marker for DCs (Fig. 2, column 2). MHC class I and II are present on both the tumor and the DCs (data not shown). Few of the DCs expressed cytokeratin and none of the tumor cells expressed CD11c (Fig. 2). The fusion efficiency was determined by two-color immunofluorescence staining, using cytokeratin and CD11c antibodies specific for the tumor cell and the DC, respectively. The fusion efficiency varied from 20 to 50% for each fusion (all fusion data not shown). MET Mice Survival Enhanced with Treatment All mice were allowed to survive at least 129 days. All mice in the control group had died by 129 days and the necessary follow-up time for the other groups was not known and may have been very long. Kaplan– Meier analysis compares survival across the whole curve. Figure 3 shows a significant difference (P ⬍ 0.01) in survival between the group treated with SEB/DC fusion or SEB alone (75% CI ⫾ 0.42) compared to control (11% CI ⫾ 0.28). MUC1 Serum Levels Correlate with Tumor Size During the experiment, MUC1 serum levels were measured at 10 weeks of age and necropsy (Figs. 4A and 4B). There was a significant difference between the SEB/fusion group and the control group at 10 weeks of
age (P ⫽ 0.04) (Fig. 4A). The sample size was too small to assess a difference in serum levels between the SEB group and the control group. At necropsy there was no difference in serum levels (Fig. 4B). However, MUC1 serum levels acquired at necropsy correlated significantly with tumor size (Pearson correlation coefficient r ⫽ 公0.86) (Fig. 4C). Immunization Elicits Mature CTLs At necropsy, CTLs were obtained from splenocytes and tested for reactivity against MUC1 expressing B16 melanoma tumor cells in a standard 8-h Cr-release assay [2, 8]. The mice treated with SEB/DC fusion as well as with SEB exhibited high levels of the MUC1specific CTLs (Fig. 5). The CTL activity in control mice, which were sacrificed at morbidity, is shown in the figure but was not included in the analysis, as CTL activity diminishes greatly at morbidity. The CTLs were detected in freshly isolated splenocytes without any in vitro manipulation, suggesting that the CTLs were mature in nature, whereas the naturally occurring CTLs in MET mice described previously were precursor CTLs [3]. Although the CTLs were cytotoxic in vitro, they apparently had little effect on the growing tumors in vivo in the MET mice. Tumors from both treated and control groups were examined using immunohistochemistry. We showed previously that the dysplastic acinar cells that had developed by 3 weeks of age expressed underglycosy-
FIG. 1. The immunization strategy and time line showing tumor progression. There was microscopic growth of pancreatic tumor when treatment began. Time points of immunologic monitoring are shown.
McCONNELL ET AL.: DC–TUMOR FUSION IMMUNOTHERAPY
199
FIG. 2. Characterization of the fusion vaccine. Flow cytometric analysis of antigens of tumor cells and DCs is represented. MET tumor cells were labeled with antibody bearing fluorescent dyes to evaluate the expression and presence of cytokeratin. Antibody to CD11c is the prominent antigen expressed on DCs. Fusion of the DCs and MET tumor was at a ratio of 5:1. DCs fused with MET tumor cells (DC:tumor fusion) were analyzed by flow cytometry. The percentage of cells expressing both cytokeratin and CD11c represents the fused cells.
lated MUC1 as determined by reactivity with the monoclonal antibodies HMFG-2, SM-3, and BC2 [2]. A similar staining pattern was observed by all three groups of mice
FIG. 3. The Kaplan–Meier survival of the mice after treatment. The Kaplan–Meier survival curve demonstrated improved survival of SEB/fusion (E) and SEB (‚) treatment mice (P ⬍ 0.01) based on a Fisher exact test.
in this study, regardless of the treatment strategy. Tumors progressed to well-differentiated acinar cell carcinomas by 15 weeks and continued to express MUC1 strongly by immunohistochemistry (data not shown). Additionally, both treated and control MET mice developed large solid tumor masses of less well-differentiated acinar cell carcinomas which failed to express human MUC1 strongly, although the cytoplasmic tail monoclonal antibody CT2, which is glycosylation insensitive and reacts with both human and mouse Muc1, showed high levels of MUC1 cytoplasmic tail expression (data not shown). There is no effective way to differentiate between the mouse and human forms of MUC1 using the cytoplasmic tail monoclonal antibody. In both treated and control mice, there were regions of MUC1 expressing tumor cells and regions of the tumor in which the tandem repeat epitopes of MUC1 appeared to be masked by glycosylation or the MUC1-expressing cells had been eliminated by the CTLs. At this time we have been unable to determine whether MUC1 epitopes have been masked by glycosylation or whether the MUC1 expressing cells have been largely eliminated. Importantly, there was
200
JOURNAL OF SURGICAL RESEARCH: VOL. 107, NO. 2, OCTOBER 2002
FIG. 5. Mature MUC1-specific CTLs were elicited by SEB/fusion and SEB treatments. CTL activity was determined with an 8-h 51 Cr-release assay using B16.MUC1 as tumor target cells. The effector:target ratio was 100:1. Both SEB/fusion and SEB mice showed high levels of mature CTL activity, without any in vitro manipulation. The low level of CTL activity in control mice was probably due to their morbid state.
mice, despite the presence of activated CTLs and improved survival. DISCUSSION
FIG. 4. MUC1 serum levels. MUC1 serum levels were determined by ELISA at 10 weeks of age and at necropsy. A. Demonstration of the correlation with SEB/fusion (E) and significantly lower MUC1 serum levels (P ⬍ 0.04) from a one-way ANOVA model with Tukey–Kramer adjustment for multiple comparisons. B. The MUC1 serum levels are not significantly different between treated and control groups at morbidity. C. The MUC1 serum levels correlate directly to the tumor size (r value ⫽ 公0.86 by Pearson correlation).
A major goal of immunotherapy is to generate tumorspecific CTLs that can effectively eliminate the growing tumor. Previous research has been focused on identifying and characterizing proteins expressed on tumor cells that may serve as potential tumor-specific antigens for recognition by CTLs [9]. Some of the most promising candidates represent conventional cellular proteins that are expressed in both normal and transformed cells [10 –21]. One such candidate is MUC1 [22, 23]. Although MUC1 is normally expressed in epithelial cells lining ducts and glands at low levels, it is a
no detectable difference in the treated groups from the control group. Tumor Size Was Not Altered with Treatment Although there was a trend toward smaller pancreatic tumor size in the treatment groups, the results did not reach statistical significance (Fig. 6). Mice were weighed and examined on a weekly basis. At necropsy, MET mice were rapidly losing weight, an indication of morbidity. Tumors were harvested and wet weight was determined. Tumors progressed rapidly in treatment
FIG. 6. Pancreas weights. Pancreas tumor weights are shown as a percentage of total body weight. Mice were sacrificed and pancreatic tumors were fixed in methacarn, air dried, and weighed. The percentage of tumor weight relative to body weight appeared larger in the control mice but was not found to be statistically significant (P ⫽ 0.67).
McCONNELL ET AL.: DC–TUMOR FUSION IMMUNOTHERAPY
target for immunotherapy, because during tumorigenesis, MUC1 is significantly altered in expression. There is an increase in the amount of MUC1 expressed on cells and in the circulation. The distribution of MUC1 is no longer restricted to the apical surface of the ducts and glands, but is found throughout the tumor mass and on the surface of tumor cells. In addition, the glycosylation of MUC1 is altered in tumor cells [24 – 26]. The oligosaccharides are shorter and fewer in number, leading to the exposure of immunodominant peptide sequences that on normal cell surfaces would be sequestered by glycosylation. The presence of MUC1 on the pancreatic tumor cells allowed us to follow the response to MUC1 following vaccination with a multivalent vaccine approach, DCs fused with tumor cells. In a previous study, we have shown that the MET mice have naturally occurring MUC1-specific precursor CTLs (pCTLs) and that the pCTLs are class I restricted and recognize the tandom repeat region of MUC1 [2, 3]. Importantly, we have shown that the CTLs can be isolated and can eradicate MUC1expressing tumor cells when adoptively transferred in vivo [3]. Although the pCTLs were present in the MET mice, the mice were not protected from the tumor, suggesting that the pCTLs were not activated enough to kill the growing tumor or that the pCTLs arose too late in tumor progression. These data suggested that immunotherapeutic regimens which stimulate MUC1specific pCTLs to become mature may be effective in decreasing tumor burden as well as enhancing survival. In addition, several investigators have shown that primary tumors treated early in their course are susceptible to immune intervention. This has relevance to several human tumors, as class I restricted CTLs have been described in humans in association with other epithelial cancers [27, 28]. In the present study, we hypothesized that specific and effective immunotherapeutic vaccines would activate the naturally existing pCTLs described previously [2] and lead to eradication of pancreatic tumor in the MET mice. Our results demonstrate that: (1) survival was significantly enhanced in vaccine as well as SEB treated MET mice (75%) compared to untreated controls (11%); (2) there was significant increase in mature CTLs in the vaccine treated as well as SEB treated MET mice; (3) MUC1 serum levels at 10 weeks of age in the vaccine treated MET mice were 50% less than that of SEB treated mice or untreated controls (P ⬍ 0.04); and (4) MUC1 serum levels significantly correlated with tumor weight at necropsy. Immunization with SEB alone can activate the pCTLs and may provide substantial help by activating other T cell repertoires needed for effective anti-tumor immune response. There was a significant survival benefit (Fig. 3) with both treatment groups and large numbers of mature CTLs were detected in immunized
201
MET mice. It is possible that the production by tumor cells of TGF, which we showed previously [2], or other immunosuppressive proteins may, in fact, suppress the antitumor effects of the mature CTLs (reviewed in [29]). In addition, it is plausible that within the tumor environment, the mature CTLs that we detected are tolerized or anergized. Understanding the mechanisms of tumor evasion is beyond the scope of this paper and more experiments are necessary to further advance this study. Our results showed a significant decrease in MUC1 serum levels early in tumor progression (10 weeks of age, when tumor burden is low) in mice treated with SEB/fusion (Fig. 4A), suggesting that the vaccination with the fused cells has an effect on tumor growth. Results also clearly indicated a direct correlation between serum MUC1 levels and tumor size (Fig. 4C). This decrease in MUC1 levels with a specific treatment (the fusion vaccine vs SEB, which is a nonspecific immune stimulant) suggests a potential clinical benefit with this approach. Clinically in humans, MUC1 serum levels have been used to track tumor progression [7, 30]. In this mouse model the difference in serum MUC1 levels or tumor weight was undetectable at necropsy (Figs. 4B and 6). One can speculate that this early benefit is lost as this aggressive oncogeneinduced tumor progresses. This is the first evidence demonstrating that mature MUC1-specific CTLs can be elicited by DC/tumor cell fusion and SEB in a spontaneous mouse model of pancreatic cancer. These CTLs were detected prior to clinical evidence of morbidity identified by rapid and acute weight loss, poor grooming, and hunched faces. Moreover, no detectable autoimmune destruction of normal tissues was observed. By following the CTL response to a tumor antigen, in our case MUC1, we should get an estimate of the general tumor-specific CTL response. We did not determine whether there was activity of the infiltrating lymphocytes within the tumors. Data from another spontaneous mouse tumor model suggest that, in the tumor environment, the CTLs may be tolerized (Mukherjee, personal communication). Future studies will address this issue in greater detail. These MET mice appropriately mimic conditions that occur in human tumors and are an excellent model for testing anti-tumor vaccine strategies. In humans, pancreatic cancer is typically found at a late stage of disease when metastatic tumor already exists. To provide an effective immune response after surgical resection of the primary tumor and prevent progression to metastatic tumor growth may be sufficient to enhance survival. Our immunization strategy does not arrest tumor growth entirely, but demonstrates a clear survival benefit in the mice. Future studies into understanding the immunosuppressive effects of these tu-
202
JOURNAL OF SURGICAL RESEARCH: VOL. 107, NO. 2, OCTOBER 2002
mors and tolerance may provide insight into designing effective immunotherapeutic strategies. 15.
ACKNOWLEDGMENTS We are grateful to Biomira, Inc., for supplying reagents and to the surgical pathologists, Drs. C. R. Conley and T. K. Lidner, for analysis of tumor samples. We thank Marvin H. Ruona for computer graphics, Jim Tarara for FACs assistance, and Carol Williams for administrative assistance and manuscript preparation.
16.
17.
REFERENCES 1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13. 14.
Rowse, G. J., Tempero, R. M., VanLith, M. L., Hollingsworth, M. A., and Gendler, S. J. Tolerance and immunity to MUC1 in a human MUC1 transgenic murine model. Cancer Res. 58: 315, 1998. Mukherjee, P., Ginardi, A. R., Madsen, C. S., Sterner, C. J., Adriance, M. C., Tevethia, M. J., and Gendler, S. J. Mice with spontaneous pancreatic cancer naturally develop MUC1specific CTLs that eradicate tumors when adoptively transferred. J. Immunol. 165(6): 3451, 2000. Mukherjee, P., Ginardi, A. R., Tinder, T. L., Sterner, C. J., and Gendler, S. J. MUC1-specific CTLs eradicate tumors when adoptively transferred in vivo. Clin. Cancer Res. 7: 848s, 2001. Rosenberg, S. A., et al. Treatment of patients with metastatic melanoma with autologous tumor-infiltrating lymphocytes and interleukin 2. J. Natl. Cancer Inst. 86(15): 1159, 1994. Tevethia, M. J., Bonneau, R. H., Griffith, J. W., and Mylin, L. A simian virus 40 large T-antigen segment containing amino acids 1 to 127 and expressed under the control of the rat elastase-1 promoter produces pancreatic acinar carcinomas in transgenic mice. J. Virol. 71(11): 8157, 1997. Gong, J., Chen, D., Kashiwaba, M., and Kufe, D. Induction of antitumor activity by immunization with fusions of dendritic and carcinoma cells. Nat. Med. 3(5): 558, 1997. Reddish, M. A., MacLean, G. D., Poppema, S., Berg, A., and Longenecker, B. M. Pre-immunotherapy serum CA27.29 (MUC-1) mucin level and CD69⫹ lymphocytes correlate with effects of Theratope sialyl-Tn-KLH cancer vaccine in active specific immunotherapy. Cancer Immunol. Immunother. 42(5): 303, 1996. Hoffman, R. W., Quinones, R. R., and Shaw, S. Anti-T3 antibody both activates and inhibits the cytotoxic activity of human T cell clones. Behring Inst. Mitt. 77: 30, 1985. Morgan, D. J., Kreuwel, H. T., Fleck, S., Levitsky, H. I., Pardoll, D. M., and Sherman, L. A. Activation of low avidity CTL specific for a self epitope results in tumor rejection but not autoimmunity. J. Immunol. 160(2): 643, 1998. Cabaniols, J. P., Cibotti, R., Kourilsky, P., Kosmatopoulos, K., and Kanellopoulos, J. M. Dose-dependent T cell tolerance to an immunodominant self peptide. Eur. J. Immunol. 24(8): 1743, 1994. Goverman, J., Woods, A., Larson, L., Weiner, L. P., Hood, L., and Zaller, D. M. Transgenic mice that express a myclin basic protein-specific T cell receptor develop spontaneous autoimmunity. Cell 72(4): 551, 1993. von Herrath, M. G., Dockter, J., and Oldstone, M. B. How virus induces a rapid or slow onset insulin-dependent diabetes mellitus in a transgenic model. Immunity 1(3): 231, 1994. Miller, J. F., and Morahan, G. Peripheral T cell tolerance. Annu. Rev. Immunol. 10: 51, 1992. Oehen, S. U., Ohashi, P. S., Burki, K., Hengartner, H., Zinkernagel, R. M., and Aichele, P. Escape of thymocytes and mature
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29. 30.
T cells from clonal deletion due to limiting tolerogen expression levels. Cell. Immunol. 158(2): 342, 1994. Ohashi, P. S., Oehen, S., Buerki, K., Pircher, H., Ohashi, C. T., Odermatt, B., Malissen, B., Zinkernagel, R. M., and Hengartner, H. Ablation of “tolerance” and induction of diabetes by virus infection in viral antigen transgenic mice. Cell 65(2): 305, 1991. Oldstone, M. B., Nerenberg, M., Southern, P., Price, J., and Lewicki, H. Virus infection triggers insulin-dependent diabetes mellitus in a transgenic model: Role of anti-self (virus) immune response. Cell. 65(2): 319, 1991. Poindexter, N. J., Landon, C., Whiteley, P. J., and Kapp, J. A. Comparison of the T cell receptors on insulin-specific hybridomas from insulin transgenic and nontransgenic mice. Loss of a subpopulation of self-reactive clones. J. Immunol. 149(1): 38, 1992. Poplonski, L., Vukusic, B., Pawling, J., Clapoff, S., Roder, J., Hozumi, N., and Wither, J. Tolerance is overcome in beef insulin-transgenic mice by activation of low-affinity autoreactive T cells. Eur. J. Immunol. 26(3): 601, 1996. Whiteley, P. J., Poindexter, N. J., Landon, C., and Kapp, J. A. A peripheral mechanism preserves self-tolerance to a secreted protein in transgenic mice. J. Immunol. 145(5): 1376, 1990. Yule, T. D., Basten, A., and Allen, P. M. Hen egg-white lysozyme-specific T cells elicited in hen egg-white lysozymetransgenic mice retain an imprint of self-tolerance. J. Immunol. 151(6): 3057, 1993. Morgan, D. J., Kreuwel, H. T., and Sherman, L. A. Antigen concentration and precursor frequency determine the rate of CD8⫹ T cell tolerance to peripherally expressed antigens. J. Immunol. 163(2): 723, 1999. Barnd, D. L., Lan, M. S., Metzgar, R. S., and Finn, O. J. Specific, major histocompatibility complex-unrestricted recognition of tumor-associated mucins by human cytotoxic T cells. Proc. Natl. Acad. Sci. USA 86(18): 7159, 1989. Ioannides, C. G., Fisk, B., Jerome, K. R., Irimura, T., Wharton, J. T., and Finn, O. J. Cytotoxic T cells from ovarian malignant tumors can recognize polymorphic epithelial mucin core peptides. J. Immunol. 151(7): 3693, 1993. Burchell, J., Gendler, S., Taylor-Papadimitriou, J., Girling, A., Lewis, A., Millis, R., and Lamport, D. Development and characterization of breast cancer reactive monoclonal antibodies directed against to the core protein of the human milk mucin. Cancer Res. 47: 5476, 1987. Girling, A., Bartkova, J., Burchell, J., Gendler, S., Gillet, C., and Taylor-Papadimitriou, J. A core protein epitope of the polymorphic epithelial mucin detected by the monoclonal antibody SM-3 is selectively exposed in a range of primary carcinomas. Int. J. Cancer 43: 1072, 1989. Burchell, J., and Taylor-Papadimitriou, J. Effect of modification of carbohydrate side chains on the reactivity of antibodies with core-protein epitopes of the MUC1 gene product. Epithelial Cell Biol. 2: 155, 1993. MacLean, G. D., Reddish, M. A., Bowen-Yacyshyn, M. B., Poppema, S., and Longenecker, B. M. Active specific immunotherapy against adenocarcinomas. Cancer Invest. 12(1): 46, 1994. Domenech, N., Henderson, R. A., and Finn, O. J. Identification of an HLA-A11-restricted epitope from the tandem repeat domain of the epithelial tumor antigen mucin. J. Immunol. 155(10): 4766, 1995. Gorelik, L., and Flavell, R. A. Transforming growth factor-beta in T-cell biology. Nat. Rev. Immunol. 2(1): 46, 2002. MacLean, G. D., Reddish, M. A., and Longenecker, B. M. Prognostic significance of preimmunotherapy serum CA27.29 (MUC-1) mucin level after active specific immunotherapy of metastatic adenocarcinoma patients. J. Immunother. 20(1): 70, 1997.